Internal thought refers to the process of directing attention away from a primary visual task to internal cognitive processing. Internal thought is a pervasive mental activity and closely related to primary task performance. As such, automatic detection of internal thought has significant potential for user modelling in intelligent interfaces, particularly for e-learning applications. Despite the close link between the eyes and the human mind, only a few studies have investigated vergence behaviour during internal thought and none has studied moment-to-moment detection of internal thought from gaze. While prior studies relied on long-term data analysis and required a large number of gaze characteristics, we describe a novel method that is computationally light-weight and that only requires eye vergence information that is readily available from binocular eye trackers. We further propose a novel paradigm to obtain ground truth internal thought annotations that exploits human blur perception. We evaluate our method for three increasingly challenging detection tasks: (1) during a controlled math-solving task, (2) during natural viewing of lecture videos, and (3) during daily activities, such as coding, browsing, and reading. Results from these evaluations demonstrate the performance and robustness of vergence-based detection of internal thought and, as such, open up new directions for research on interfaces that adapt to shifts of mental attention.

Recent methods to automatically calibrate stationary eye trackers were shown to effectively reduce inherent calibration distortion. However, these methods require additional information, such as mouse clicks or on-screen content. We propose the first method that only requires users' eye movements to reduce calibration distortion in the background while users naturally look at an interface. Our method exploits that calibration distortion makes straight saccade trajectories appear curved between the saccadic start and end points. We show that this curving effect is systematic and the result of distorted gaze projection plane. To mitigate calibration distortion, our method undistorts this plane by straightening saccade trajectories using image warping. We show that this approach improves over the common six-point calibration and is promising for reducing distortion. As such, it provides a non-intrusive solution to alleviating accuracy decrease of eye tracker during long-term use.

While great progress has been made recently in automatic image manipulation, it has been limited to object centric images like faces or structured scene datasets. In this work, we take a step towards general scene-level image editing by developing an automatic interaction-free object removal model. Our model learns to find and remove objects from general scene images using image-level labels and unpaired data in a generative adversarial network (GAN) framework. We achieve this with two key contributions: a two-stage editor architecture consisting of a mask generator and image in-painter that co-operate to remove objects, and a novel GAN based prior for the mask generator that allows us to flexibly incorporate knowledge about object shapes. We experimentally show on two datasets that our method effectively removes a wide variety of objects using weak supervision only

Deep neural perception and control networks have become key com- ponents of self-driving vehicles. User acceptance is likely to benefit from easy- to-interpret textual explanations which allow end-users to understand what trig- gered a particular behavior. Explanations may be triggered by the neural con- troller, namely introspective explanations , or informed by the neural controller’s output, namely rationalizations . We propose a new approach to introspective ex- planations which consists of two parts. First, we use a visual (spatial) attention model to train a convolutional network end-to-end from images to the vehicle control commands, i . e ., acceleration and change of course. The controller’s at- tention identifies image regions that potentially influence the network’s output. Second, we use an attention-based video-to-text model to produce textual ex- planations of model actions. The attention maps of controller and explanation model are aligned so that explanations are grounded in the parts of the scene that mattered to the controller. We explore two approaches to attention alignment, strong- and weak-alignment. Finally, we explore a version of our model that generates rationalizations, and compare with introspective explanations on the same video segments. We evaluate these models on a novel driving dataset with ground-truth human explanations, the Berkeley DeepDrive eXplanation (BDD- X) dataset. Code is available at https://github.com/JinkyuKimUCB/explainable-deep-driving

Due to the importance of zero-shot learning, i.e. classifying images where there is a lack of labeled training data, the number of proposed approaches has recently increased steadily. We argue that it is time to take a step back and to analyze the status quo of the area. The purpose of this paper is three-fold. First, given the fact that there is no agreed upon zero-shot learning benchmark, we first define a new benchmark by unifying both the evaluation protocols and data splits of publicly available datasets used for this task. This is an important contribution as published results are often not comparable and sometimes even flawed due to, e.g. pre-training on zero-shot test classes. Moreover, we propose a new zero-shot learning dataset, the Animals with Attributes 2 (AWA2) dataset which we make publicly available both in terms of image features and the images themselves. Second, we compare and analyze a significant number of the state-of-the-art methods in depth, both in the classic zero-shot setting but also in the more realistic generalized zero-shot setting. Finally, we discuss in detail the limitations of the current status of the area which can be taken as a basis for advancing it.

Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods and the “perfect single frame detector”. We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech pedestrian dataset). After manually clustering the frequent errors of a top detector, we characterise both localisation and background- versus-foreground errors. To address localisation errors we study the impact of training annotation noise on the detector performance, and show that we can improve results even with a small portion of sanitised training data. To address background/foreground discrimination, we study convnets for pedestrian detection, and discuss which factors affect their performance. Other than our in-depth analysis, we report top performance on the Caltech pedestrian dataset, and provide a new sanitised set of training and test annotations.

Natural language explanations of deep neural network decisions provide an intuitive way for a AI agent to articulate a reasoning process. Current textual explanations learn to discuss class discriminative features in an image. However, it is also helpful to understand which attributes might change a classification decision if present in an image (e.g., "This is not a Scarlet Tanager because it does not have black wings.") We call such textual explanations counterfactual explanations, and propose an intuitive method to generate counterfactual explanations by inspecting which evidence in an input is missing, but might contribute to a different classification decision if present in the image. To demonstrate our method we consider a fine-grained image classification task in which we take as input an image and a counterfactual class and output text which explains why the image does not belong to a counterfactual class. We then analyze our generated counterfactual explanations both qualitatively and quantitatively using proposed automatic metrics.

The inner structure of a material is called microstructure. It stores the genesis of a material and determines all its physical and chemical properties. While microstructural characterization is widely spread and well known, the microstructural classification is mostly done manually by human experts, which opens doors for huge uncertainties. Since the microstructure could be a combination of different phases with complex substructures its automatic classification is very challenging and just a little work in this field has been carried out. Prior related works apply mostly designed and engineered features by experts and classify microstructure separately from feature extraction step. Recently Deep Learning methods have shown surprisingly good performance in vision applications by learning the features from data together with the classification step. In this work, we propose a deep learning method for microstructure classification in the examples of certain microstructural constituents of low carbon steel. This novel method employs pixel-wise segmentation via Fully Convolutional Neural Networks (FCNN) accompanied by max-voting scheme. Our system achieves 93.94% classification accuracy, drastically outperforming the state-of-the-art method of 48.89% accuracy, indicating the effectiveness of pixel-wise approaches. Beyond the success presented in this paper, this line of research offers a more robust and first of all objective way for the difficult task of steel quality appreciation.

In-depth scene descriptions and question answering tasks have greatly increased the scope of today's definition of scene understanding. While such tasks are in principle open ended, current formulations primarily focus on describing only the current state of the scenes under consideration. In contrast, in this paper, we focus on the future states of the scenes which are also conditioned on actions. We posit this as a question answering task, where an answer has to be given about a future scene state, given observations of the current scene, and a question that includes a hypothetical action. Our solution is a hybrid model which integrates a physics engine into a question answering architecture in order to anticipate future scene states resulting from object-object interactions caused by an action. We demonstrate first results on this challenging new problem and compare to baselines, where we outperform fully data-driven end-to-end learning approaches.

The matching of multiple objects (e.g. shapes or images) is a fundamental problem in vision and graphics. In order to robustly handle ambiguities, noise and repetitive patterns in challenging real-world settings, it is essential to take geometric consistency between points into account. Computationally, the multi-matching problem is difficult. It can be phrased as simultaneously solving multiple (NP-hard) quadratic assignment problems (QAPs) that are coupled via cycle-consistency constraints. The main limitations of existing multi-matching methods are that they either ignore geometric consistency and thus have limited robustness, or they are restricted to small-scale problems due to their (relatively) high computational cost. We address these shortcomings by introducing a Higher-order Projected Power Iteration method, which is (i) efficient and scales to tens of thousands of points, (ii) straightforward to implement, (iii) able to incorporate geometric consistency, and (iv) guarantees cycle-consistent multi-matchings. Experimentally we show that our approach is superior to existing methods.

For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence. This problem can be formalized as a sequence prediction problem, where a number of observations are used to predict the sequence into the future. However, real-world scenarios demand a model of uncertainty of such predictions, as future states become increasingly uncertain and multi-modal -- in particular on long time horizons. This makes modelling and learning challenging. We cast state of the art semantic segmentation and future prediction models based on deep learning into a Bayesian formulation that in turn allows for a full Bayesian treatment of the prediction problem. We present a new sampling scheme for this model that draws from the success of variational autoencoders by incorporating a recognition network. In the experiments we show that our model outperforms prior work in accuracy of the predicted segmentation and provides calibrated probabilities that also better capture the multi-modal aspects of possible future states of street scenes.

We introduce Primal-Dual Wasserstein GAN, a new learning algorithm for building latent variable models of the data distribution based on the primal and the dual formulations of the optimal transport (OT) problem. We utilize the primal formulation to learn a flexible inference mechanism and to create an optimal approximate coupling between the data distribution and the generative model. In order to learn the generative model, we use the dual formulation and train the decoder adversarially through a critic network that is regularized by the approximate coupling obtained from the primal. Unlike previous methods that violate various properties of the optimal critic, we regularize the norm and the direction of the gradients of the critic function. Our model shares many of the desirable properties of auto-encoding models in terms of mode coverage and latent structure, while avoiding their undesirable averaging properties, e.g. their inability to capture sharp visual features when modeling real images. We compare our algorithm with several other generative modeling techniques that utilize Wasserstein distances on Frechet Inception Distance (FID) and Inception Scores (IS).

With the widespread use of machine learning (ML) techniques, ML as a service has become increasingly popular. In this setting, an ML model resides on a server and users can query the model with their data via an API. However, if the user's input is sensitive, sending it to the server is not an option. Equally, the service provider does not want to share the model by sending it to the client for protecting its intellectual property and pay-per-query business model. In this paper, we propose MLCapsule, a guarded offline deployment of machine learning as a service. MLCapsule executes the machine learning model locally on the user's client and therefore the data never leaves the client. Meanwhile, MLCapsule offers the service provider the same level of control and security of its model as the commonly used server-side execution. In addition, MLCapsule is applicable to offline applications that require local execution. Beyond protecting against direct model access, we demonstrate that MLCapsule allows for implementing defenses against advanced attacks on machine learning models such as model stealing/reverse engineering and membership inference.

In this study, we explore building a two-stage framework for enabling users to directly manipulate high-level attributes of a natural scene. The key to our approach is a deep generative network which can hallucinate images of a scene as if they were taken at a different season (e.g. during winter), weather condition (e.g. in a cloudy day) or time of the day (e.g. at sunset). Once the scene is hallucinated with the given attributes, the corresponding look is then transferred to the input image while preserving the semantic details intact, giving a photo-realistic manipulation result. As the proposed framework hallucinates what the scene will look like, it does not require any reference style image as commonly utilized in most of the appearance or style transfer approaches. Moreover, it allows to simultaneously manipulate a given scene according to a diverse set of transient attributes within a single model, eliminating the need of training multiple networks per each translation task. Our comprehensive set of qualitative and quantitative results demonstrate the effectiveness of our approach against the competing methods.

In combinatorial optimization, partial variable assignments are called persistent if they agree with some optimal solution. We propose persistency criteria for the multicut and max-cut problem as well as fast combinatorial routines to verify them. The criteria that we derive are based on mappings that improve feasible multicuts, respectively cuts. Our elementary criteria can be checked enumeratively. The more advanced ones rely on fast algorithms for upper and lower bounds for the respective cut problems and max-flow techniques for auxiliary min-cut problems. Our methods can be used as a preprocessing technique for reducing problem sizes or for computing partial optimality guarantees for solutions output by heuristic solvers. We show the efficacy of our methods on instances of both problems from computer vision, biomedical image analysis and statistical physics.

We propose a novel approach to jointly perform 3D object retrieval and pose estimation from monocular images.In order to make the method robust to real world scene variations in the images, e.g. texture, lighting and background,we learn an embedding space from 3D data that only includes the relevant information, namely the shape and pose.Our method can then be trained for robustness under real world scene variations without having to render a large training set simulating these variations. Our learned embedding explicitly disentangles a shape vector and a pose vector, which alleviates both pose bias for 3D shape retrieval and categorical bias for pose estimation. Having the learned disentangled embedding, we train a CNN to map the images to the embedding space, and then retrieve the closest 3D shape from the database and estimate the 6D pose of the object using the embedding vectors. Our method achieves 10.8 median error for pose estimation and 0.514 top-1-accuracy for category agnostic 3D object retrieval on the Pascal3D+ dataset. It therefore outperforms the previous state-of-the-art methods on both tasks.

From autonomous driving cars to surgical robots, robotic system has enjoyed significant growth over the past decade. With the rapid development in robotics alongside the evolution in the related fields, such as computer vision and machine learning, integrating perception, anticipation and manipulation is key to the success of future robotic system. In this thesis, we explore different ways of such integration to extend the capabilities of a robotic system to take on more challenging real world tasks. On anticipation and perception, we address the recognition of ongoing activity from videos. In particular we focus on long-duration and complex activities and hence propose a new challenging dataset to facilitate the work. We introduce hierarchical labels over the activity classes and investigate the temporal accuracy-specificity trade-offs. We propose a new method based on recurrent neural networks that learns to predict over this hierarchy and realize accuracy specificity trade-offs. Our method outperforms several baselines on this new challenge. On manipulation with perception, we propose an efficient framework for programming a robot to use human tools. We first present a novel and compact model for using tools described by a tip model. Then we explore a strategy of utilizing a dual-gripper approach for manipulating tools – motivated by the absence of dexterous hands on widely available general purpose robots. Afterwards, we embed the tool use learning into a hierarchical architecture and evaluate it on a Baxter research robot. Finally, combining perception, anticipation and manipulation, we focus on a block stacking task. First we explore how to guide robot to place a single block into the scene without collapsing the existing structure. We introduce a mechanism to predict physical stability directly from visual input and evaluate it first on a synthetic data and then on real-world block stacking. Further, we introduce the target stacking task where the agent stacks blocks to reproduce a tower shown in an image. To do so, we create a synthetic block stacking environment with physics simulation in which the agent can learn block stacking end-to-end through trial and error, bypassing to explicitly model the corresponding physics knowledge. We propose a goal-parametrized GDQN model to plan with respect to the specific goal. We validate the model on both a navigation task in a classic gridworld environment and the block stacking task.

We propose a deep representation of appearance, i. e. the relation of color, surface orientation, viewer position, material and illumination. Previous approaches have used deep learning to extract classic appearance representations relating to reflectance model parameters (e. g. Phong) or illumination (e. g. HDR environment maps). We suggest to directly represent appearance itself as a network we call a deep appearance map (DAM). This is a 4D generalization over 2D reflectance maps, which held the view direction fixed. First, we show how a DAM can be learned from images or video frames and later be used to synthesize appearance, given new surface orientations and viewer positions. Second, we demonstrate how another network can be used to map from an image or video frames to a DAM network to reproduce this appearance, without using a lengthy optimization such as stochastic gradient descent (learning-to-learn). Finally, we generalize this to an appearance estimation-and-segmentation task, where we map from an image showing multiple materials to multiple networks reproducing their appearance, as well as per-pixel segmentation.

Machine learning is transforming the world. Its application areas span privacy sensitive and security critical tasks such as human identification and self-driving cars. These applications raise privacy and security related questions that are not fully understood or answered yet: Can automatic person recognisers identify people in photos even when their faces are blurred? How easy is it to find an adversarial input for a self-driving car that makes it drive off the road? This thesis contributes one of the first steps towards a better understanding of such concerns. We observe that many privacy and security critical scenarios for learned models involve input data manipulation: users obfuscate their identity by blurring their faces and adversaries inject imperceptible perturbations to the input signal. We introduce a data manipulator framework as a tool for collectively describing and analysing privacy and security relevant scenarios involving learned models. A data manipulator introduces a shift in data distribution for achieving privacy or security related goals, and feeds the transformed input to the target model. This framework provides a common perspective on the studies presented in the thesis. We begin the studies from the user’s privacy point of view. We analyse the efficacy of common obfuscation methods like face blurring, and show that they are surprisingly ineffective against state of the art person recognition systems. We then propose alternatives based on head inpainting and adversarial examples. By studying the user privacy, we also study the dual problem: model security. In model security perspective, a model ought to be robust and reliable against small amounts of data manipulation. In both cases, data are manipulated with the goal of changing the target model prediction. User privacy and model security problems can be described with the same objective. We then study the knowledge aspect of the data manipulation problem. The more one knows about the target model, the more effective manipulations one can craft. We propose a game theoretic manipulation framework to systematically represent the knowledge level on the target model and derive privacy and security guarantees. We then discuss ways to increase knowledge about a black-box model by only querying it, deriving implications that are relevant to both privacy and security perspectives.

Machine Learning techniques are widely used by online services (e.g. Google, Apple) in order to analyze and make predictions on user data. As many of the provided services are user-centric (e.g. personal photo collections, speech recognition, personal assistance), user data generated on personal devices is key to provide the service. In order to protect the data and the privacy of the user, federated learning techniques have been proposed where the data never leaves the user's device and "only" model updates are communicated back to the server. In our work, we propose a new threat model that is not concerned with learning about the content - but rather is concerned with the linkability of users during such decentralized learning scenarios. We show that model updates are characteristic for users and therefore lend themselves to linkability attacks. We show identification and matching of users across devices in closed and open world scenarios. In our experiments, we find our attacks to be highly effective, achieving 20x-175x chance-level performance. In order to mitigate the risks of linkability attacks, we study various strategies. As adding random noise does not offer convincing operation points, we propose strategies based on using calibrated domain-specific data; we find these strategies offers substantial protection against linkability threats with little effect to utility.

Machine Learning (ML) models are increasingly deployed in the wild to perform a wide range of tasks. In this work, we ask to what extent can an adversary steal functionality of such "victim" models based solely on blackbox interactions: image in, predictions out. In contrast to prior work, we present an adversary lacking knowledge of train/test data used by the model, its internals, and semantics over model outputs. We formulate model functionality stealing as a two-step approach: (i) querying a set of input images to the blackbox model to obtain predictions; and (ii) training a "knockoff" with queried image-prediction pairs. We make multiple remarkable observations: (a) querying random images from a different distribution than that of the blackbox training data results in a well-performing knockoff; (b) this is possible even when the knockoff is represented using a different architecture; and (c) our reinforcement learning approach additionally improves query sample efficiency in certain settings and provides performance gains. We validate model functionality stealing on a range of datasets and tasks, as well as on a popular image analysis API where we create a reasonable knockoff for as little as $30.

Importance of visual context in scene understanding tasks is well recognized in the computer vision community. However, to what extent the computer vision models for image classification and semantic segmentation are dependent on the context to make their predictions is unclear. A model overly relying on context will fail when encountering objects in context distributions different from training data and hence it is important to identify these dependencies before we can deploy the models in the real-world. We propose a method to quantify the sensitivity of black-box vision models to visual context by editing images to remove selected objects and measuring the response of the target models. We apply this methodology on two tasks, image classification and semantic segmentation, and discover undesirable dependency between objects and context, for example that "sidewalk" segmentation relies heavily on "cars" being present in the image. We propose an object removal based data augmentation solution to mitigate this dependency and increase the robustness of classification and segmentation models to contextual variations. Our experiments show that the proposed data augmentation helps these models improve the performance in out-of-context scenarios, while preserving the performance on regular data.

As first-person cameras in head-mounted displays become increasingly prevalent, so does the problem of infringing user and bystander privacy. To address this challenge, we present PrivacEye, a proof-of-concept system that detects privacysensitive everyday situations and automatically enables and disables the first-person camera using a mechanical shutter. To close the shutter, PrivacEye detects sensitive situations from first-person camera videos using an end-to-end deep-learning model. To open the shutter without visual input, PrivacEye uses a separate, smaller eye camera to detect changes in users' eye movements to gauge changes in the "privacy level" of the current situation. We evaluate PrivacEye on a dataset of first-person videos recorded in the daily life of 17 participants that they annotated with privacy sensitivity levels. We discuss the strengths and weaknesses of our proof-of-concept system based on a quantitative technical evaluation as well as qualitative insights from semi-structured interviews.

Obtaining deep networks that are robust against adversarial examples and generalize well is an open problem. A recent hypothesis even states that both robust and accurate models are impossible, i.e., adversarial robustness and generalization are conflicting goals. In an effort to clarify the relationship between robustness and generalization, we assume an underlying, low-dimensional data manifold and show that: 1. regular adversarial examples leave the manifold; 2. adversarial examples constrained to the manifold, i.e., on-manifold adversarial examples, exist; 3. on-manifold adversarial examples are generalization errors, and on-manifold adversarial training boosts generalization; 4. and regular robustness is independent of generalization. These assumptions imply that both robust and accurate models are possible. However, different models (architectures, training strategies etc.) can exhibit different robustness and generalization characteristics. To confirm our claims, we present extensive experiments on synthetic data (with access to the true manifold) as well as on EMNIST, Fashion-MNIST and CelebA.

Research in computer graphics has been in pursuit of realistic image generation for a long time. Recent advances in machine learning with deep generative models have shown increasing success of closing the realism gap by using data-driven and learned components. There is an increasing concern that real and fake images will become more and more difficult to tell apart. We take a first step towards this larger research challenge by asking the question if and to what extend a generated fake image can be attribute to a particular Generative Adversarial Networks (GANs) of a certain architecture and trained with particular data and random seed. Our analysis shows single samples from GANs carry highly characteristic fingerprints which make attribution of images to GANs possible. Surprisingly, this is even possible for GANs with same architecture and same training that only differ by the training seed.

Convnets have enabled significant progress in pedestrian detection recently, but there are still open questions regarding suitable architectures and training data. We revisit CNN design and point out key adaptations, enabling plain FasterRCNN to obtain state-of-the-art results on the Caltech dataset. To achieve further improvement from more and better data, we introduce CityPersons, a new set of person annotations on top of the Cityscapes dataset. The diversity of CityPersons allows us for the first time to train one single CNN model that generalizes well over multiple benchmarks. Moreover, with additional training with CityPersons, we obtain top results using FasterRCNN on Caltech, improving especially for more difficult cases (heavy occlusion and small scale) and providing higher localization quality.

Previous work focused on predicting visual search targets from human fixations but, in the real world, a specific target is often not known, e.g. when searching for a present for a friend. In this work we instead study the problem of predicting the mental picture, i.e. only an abstract idea instead of a specific target. This task is significantly more challenging given that mental pictures of the same target category can vary widely depending on personal biases, and given that characteristic target attributes can often not be verbalised explicitly. We instead propose to use gaze information as implicit information on users' mental picture and present a novel gaze pooling layer to seamlessly integrate semantic and localized fixation information into a deep image representation. We show that we can robustly predict both the mental picture's category as well as attributes on a novel dataset containing fixation data of 14 users searching for targets on a subset of the DeepFahion dataset. Our results have important implications for future search interfaces and suggest deep gaze pooling as a general-purpose approach for gaze-supported computer vision systems.

Audio Description (AD) provides linguistic descriptions of movies and allows visually impaired people to follow a movie along with their peers. Such descriptions are by design mainly visual and thus naturally form an interesting data source for computer vision and computational linguistics. In this work we propose a novel dataset which contains transcribed ADs, which are temporally aligned to full length movies. In addition we also collected and aligned movie scripts used in prior work and compare the two sources of descriptions. In total the Large Scale Movie Description Challenge (LSMDC) contains a parallel corpus of 118,114 sentences and video clips from 202 movies. First we characterize the dataset by benchmarking different approaches for generating video descriptions. Comparing ADs to scripts, we find that ADs are indeed more visual and describe precisely what is shown rather than what should happen according to the scripts created prior to movie production. Furthermore, we present and compare the results of several teams who participated in a challenge organized in the context of the workshop "Describing and Understanding Video & The Large Scale Movie Description Challenge (LSMDC)", at ICCV 2015.

Visual object detection has seen substantial improvements during the last years due to the possibilities enabled by deep learning. While research on image classification provides continuous progress on how to learn image representations and classifiers jointly, object detection research focuses on identifying how to properly use deep learning technology to effectively localise objects. In this thesis, we analyse and improve different aspects of the commonly used detection pipeline. We analyse ten years of research on pedestrian detection and find that improvement of feature representations was the driving factor. Motivated by this finding, we adapt an end-to-end learned detector architecture from general object detection to pedestrian detection. Our deep network outperforms all previous neural networks for pedestrian detection by a large margin, even without using additional training data. After substantial improvements on pedestrian detection in recent years, we investigate the gap between human performance and state-of-the-art pedestrian detectors. We find that pedestrian detectors still have a long way to go before they reach human performance, and we diagnose failure modes of several top performing detectors, giving direction to future research. As a side-effect we publish new, better localised annotations for the Caltech pedestrian benchmark. We analyse detection proposals as a preprocessing step for object detectors. We establish different metrics and compare a wide range of methods according to these metrics. By examining the relationship between localisation of proposals and final object detection performance, we define and experimentally verify a metric that can be used as a proxy for detector performance. Furthermore, we address a structural weakness of virtually all object detection pipelines: non-maximum suppression. We analyse why it is necessary and what the shortcomings of the most common approach are. To address these problems, we present work to overcome these shortcomings and to replace typical non-maximum suppression with a learnable alternative. The introduced paradigm paves the way to true end-to-end learning of object detectors without any post-processing. In summary, this thesis provides analyses of recent pedestrian detectors and detection proposals, improves pedestrian detection by employing deep neural networks, and presents a viable alternative to traditional non-maximum suppression.

Much progress has been made in image and video segmentation over the last years. To a large extent, the success can be attributed to the strong appearance models completely learned from data, in particular using deep learning methods. However,to perform best these methods require large representative datasets for training with expensive pixel-level annotations, which in case of videos are prohibitive to obtain. Therefore, there is a need to relax this constraint and to consider alternative forms of supervision, which are easier and cheaper to collect. In this thesis, we aim to develop algorithms for learning to segment in images and videos with different levels of supervision. First, we develop approaches for training convolutional networks with weaker forms of supervision, such as bounding boxes or image labels, for object boundary estimation and semantic/instance labelling tasks. We propose to generate pixel-level approximate groundtruth from these weaker forms of annotations to train a network, which allows to achieve high-quality results comparable to the full supervision quality without any modifications of the network architecture or the training procedure. Second, we address the problem of the excessive computational and memory costs inherent to solving video segmentation via graphs. We propose approaches to improve the runtime and memory efficiency as well as the output segmentation quality by learning from the available training data the best representation of the graph. In particular, we contribute with learning must-link constraints, the topology and edge weights of the graph as well as enhancing the graph nodes - superpixels - themselves. Third, we tackle the task of pixel-level object tracking and address the problem of the limited amount of densely annotated video data for training convolutional networks. We introduce an architecture which allows training with static images only and propose an elaborate data synthesis scheme which creates a large number of training examples close to the target domain from the given first frame mask. With the proposed techniques we show that densely annotated consequent video data is not necessary to achieve high-quality temporally coherent video segmentationresults. In summary, this thesis advances the state of the art in weakly supervised image segmentation, graph-based video segmentation and pixel-level object tracking and contributes with the new ways of training convolutional networks with a limited amount of pixel-level annotated training data.

Convolutional networks reach top quality in pixel-level object tracking but require a large amount of training data (1k ~ 10k) to deliver such results. We propose a new training strategy which achieves state-of-the-art results across three evaluation datasets while using 20x ~ 100x less annotated data than competing methods. Instead of using large training sets hoping to generalize across domains, we generate in-domain training data using the provided annotation on the first frame of each video to synthesize ("lucid dream") plausible future video frames. In-domain per-video training data allows us to train high quality appearance- and motion-based models, as well as tune the post-processing stage. This approach allows to reach competitive results even when training from only a single annotated frame, without ImageNet pre-training. Our results indicate that using a larger training set is not automatically better, and that for the tracking task a smaller training set that is closer to the target domain is more effective. This changes the mindset regarding how many training samples and general "objectness" knowledge are required for the object tracking task.

We study the problem of decomposing (clustering) a tree with respect to costs attributed to pairs of nodes, so as to minimize the sum of costs for those pairs of nodes that are in the same component (cluster). For the general case and for the special case of the tree being a star, we show that the problem is NP-hard. For the special case of the tree being a path, this problem is known to be polynomial time solvable. We characterize several classes of facets of the combinatorial polytope associated with a formulation of this clustering problem in terms of lifted multicuts. In particular, our results yield a complete totally dual integral (TDI) description of the lifted multicut polytope for paths, which establishes a connection to the combinatorial properties of alternative formulations such as set partitioning.

Modern image classification methods are based on supervised learning algorithms that require labeled training data. However, only a limited amount of annotated data may be available in certain applications due to scarcity of the data itself or high costs associated with human annotation. Introduction of additional information and structural constraints can help improve the performance of a learning algorithm. In this thesis, we study the framework of learning using privileged information and demonstrate its relation to learning with instance weights. We also consider multitask feature learning and develop an efficient dual optimization scheme that is particularly well suited to problems with high dimensional image descriptors. Scaling annotation to a large number of image categories leads to the problem of class ambiguity where clear distinction between the classes is no longer possible. Many real world images are naturally multilabel yet the existing annotation might only contain a single label. In this thesis, we propose and analyze a number of loss functions that allow for a certain tolerance in top k predictions of a learner. Our results indicate consistent improvements over the standard loss functions that put more penalty on the first incorrect prediction compared to the proposed losses. All proposed learning methods are complemented with efficient optimization schemes that are based on stochastic dual coordinate ascent for convex problems and on gradient descent for nonconvex formulations.

Understanding physical phenomena is a key component of human intelligence and enables physical interaction with previously unseen environments. In this paper, we study how an artificial agent can autonomously acquire this intuition through interaction with the environment. We created a synthetic block stacking environment with physics simulation in which the agent can learn a policy end-to-end through trial and error. Thereby, we bypass to explicitly model physical knowledge within the policy. We are specifically interested in tasks that require the agent to reach a given goal state that may be different for every new trial. To this end, we propose a deep reinforcement learning framework that learns policies which are parametrized by a goal. We validated the model on a toy example navigating in a grid world with different target positions and in a block stacking task with different target structures of the final tower. In contrast to prior work, our policies show better generalization across different goals.

Computer Vision has undergone major changes over the recent five years. Here, we investigate if the performance of such architectures generalizes to more complex tasks that require a more holistic approach to scene comprehension. The presented work focuses on learning spatial and multi-modal representations, and the foundations of a Visual Turing Test, where the scene understanding is tested by a series of questions about its content. In our studies, we propose DAQUAR, the first ‘question answering about real-world images’ dataset together with methods, termed a symbolic-based and a neural-based visual question answering architectures, that address the problem. The symbolic-based method relies on a semantic parser, a database of visual facts, and a bayesian formulation that accounts for various interpretations of the visual scene. The neural-based method is an end-to-end architecture composed of a question encoder, image encoder, multimodal embedding, and answer decoder. This architecture has proven to be effective in capturing language-based biases. It also becomes the standard component of other visual question answering architectures. Along with the methods, we also investigate various evaluation metrics that embraces uncertainty in word's meaning, and various interpretations of the scene and the question.

People nowadays share large parts of their personal lives through social media. Being able to automatically recognise people in personal photos may greatly enhance user convenience by easing photo album organisation. For human identification task, however, traditional focus of computer vision has been face recognition and pedestrian re-identification. Person recognition in social media photos sets new challenges for computer vision, including non-cooperative subjects (e.g. backward viewpoints, unusual poses) and great changes in appearance. To tackle this problem, we build a simple person recognition framework that leverages convnet features from multiple image regions (head, body, etc.). We propose new recognition scenarios that focus on the time and appearance gap between training and testing samples. We present an in-depth analysis of the importance of different features according to time and viewpoint generalisability. In the process, we verify that our simple approach achieves the state of the art result on the PIPA benchmark, arguably the largest social media based benchmark for person recognition to date with diverse poses, viewpoints, social groups, and events. Compared the conference version of the paper, this paper additionally presents (1) analysis of a face recogniser (DeepID2+), (2) new method naeil2 that combines the conference version method naeil and DeepID2+ to achieve state of the art results even compared to post-conference works, (3) discussion of related work since the conference version, (4) additional analysis including the head viewpoint-wise breakdown of performance, and (5) results on the open-world setup.

Many deployed learned models are black boxes: given input, returns output. Internal information about the model, such as the architecture, optimisation procedure, or training data, is not disclosed explicitly as it might contain proprietary information or make the system more vulnerable. This work shows that such attributes of neural networks can be exposed from a sequence of queries. This has multiple implications. On the one hand, our work exposes the vulnerability of black-box neural networks to different types of attacks -- we show that the revealed internal information helps generate more effective adversarial examples against the black box model. On the other hand, this technique can be used for better protection of private content from automatic recognition models using adversarial examples. Our paper suggests that it is actually hard to draw a line between white box and black box models.

Deep models are the defacto standard in visual decision problems due to their impressive performance on a wide array of visual tasks. On the other hand, their opaqueness has led to a surge of interest in explainable systems. In this work, we emphasize the importance of model explanation in various forms such as visual pointing and textual justification. The lack of data with justification annotations is one of the bottlenecks of generating multimodal explanations. Thus, we propose two large-scale datasets with annotations that visually and textually justify a classification decision for various activities, i.e. ACT-X, and for question answering, i.e. VQA-X. We also introduce a multimodal methodology for generating visual and textual explanations simultaneously. We quantitatively show that training with the textual explanations not only yields better textual justification models, but also models that better localize the evidence that support their decision.

Generating natural language descriptions for visual data links computer vision and computational linguistics. Being able to generate a concise and human-readable description of a video is a step towards visual understanding. At the same time, grounding natural language in visual data provides disambiguation for the linguistic concepts, necessary for many applications. This thesis focuses on both directions and tackles three specific problems. First, we develop recognition approaches to understand video of complex cooking activities. We propose an approach to generate coherent multi-sentence descriptions for our videos. Furthermore, we tackle the new task of describing videos at variable level of detail. Second, we present a large-scale dataset of movies and aligned professional descriptions. We propose an approach, which learns from videos and sentences to describe movie clips relying on robust recognition of visual semantic concepts. Third, we propose an approach to ground textual phrases in images with little or no localization supervision, which we further improve by introducing Multimodal Compact Bilinear Pooling for combining language and vision representations. Finally, we jointly address the task of describing videos and grounding the described people. To summarize, this thesis advances the state-of-the-art in automatic video description and visual grounding and also contributes large datasets for studying the intersection of computer vision and computational linguistics.

What does human gaze reveal about a users' intents and to which extend can these intents be inferred or even visualized? Gaze was proposed as an implicit source of information to predict the target of visual search and, more recently, to predict the object class and attributes of the search target. In this work, we go one step further and investigate the feasibility of combining recent advances in encoding human gaze information using deep convolutional neural networks with the power of generative image models to visually decode, i.e. create a visual representation of, the search target. Such visual decoding is challenging for two reasons: 1) the search target only resides in the user's mind as a subjective visual pattern, and can most often not even be described verbally by the person, and 2) it is, as of yet, unclear if gaze fixations contain sufficient information for this task at all. We show, for the first time, that visual representations of search targets can indeed be decoded only from human gaze fixations. We propose to first encode fixations into a semantic representation and then decode this representation into an image. We evaluate our method on a recent gaze dataset of 14 participants searching for clothing in image collages and validate the model's predictions using two human studies. Our results show that 62% (Chance level = 10%) of the time users were able to select the categories of the decoded image right. In our second studies we show the importance of a local gaze encoding for decoding visual search targets of user

People are often a central element of visual scenes, particularly in real-world street scenes. Thus it has been a long-standing goal in Computer Vision to develop methods aiming at analyzing humans in visual data. Due to the complexity of real-world scenes, visual understanding of people remains challenging for machine perception. In this thesis we focus on advancing the techniques for people detection and tracking in crowded street scenes. We also propose new models for human pose estimation and motion segmentation in realistic images and videos. First, we propose detection models that are jointly trained to detect single person as well as pairs of people under varying degrees of occlusion. The learning algorithm of our joint detector facilitates a tight integration of tracking and detection, because it is designed to address common failure cases during tracking due to long-term inter-object occlusions. Second, we propose novel multi person tracking models that formulate tracking as a graph partitioning problem. Our models jointly cluster detection hypotheses in space and time, eliminating the need for a heuristic non-maximum suppression. Furthermore, for crowded scenes, our tracking model encodes long-range person re-identification information into the detection clustering process in a unified and rigorous manner. Third, we explore the visual tracking task in different granularity. We present a tracking model that simultaneously clusters object bounding boxes and pixel level trajectories over time. This approach provides a rich understanding of the motion of objects in the scene. Last, we extend our tracking model for the multi person pose estimation task. We introduce a joint subset partitioning and labelling model where we simultaneously estimate the poses of all the people in the scene. In summary, this thesis addresses a number of diverse tasks that aim to enable vision systems to analyze people in realistic images and videos. In particular, the thesis proposes several novel ideas and rigorous mathematical formulations, pushes the boundary of state-of-the-arts and results in superior performance.

State-of-the-art learning based boundary detection methods require extensive training data. Since labelling object boundaries is one of the most expensive types of annotations, there is a need to relax the requirement to carefully annotate images to make both the training more affordable and to extend the amount of training data. In this paper we propose a technique to generate weakly supervised annotations and show that bounding box annotations alone suffice to reach high-quality object boundaries without using any object-specific boundary annotations. With the proposed weak supervision techniques we achieve the top performance on the object boundary detection task, outperforming by a large margin the current fully supervised state-of-the-art methods.

Undoing the image formation process and therefore decomposing appearance into its intrinsic properties is a challenging task due to the under-constraint nature of this inverse problem. While significant progress has been made on inferring shape, materials and illumination from images only, progress in an unconstrained setting is still limited. We propose a convolutional neural architecture to estimate reflectance maps of specular materials in natural lighting conditions. We achieve this in an end-to-end learning formulation that directly predicts a reflectance map from the image itself. We show how to improve estimates by facilitating additional supervision in an indirect scheme that first predicts surface orientation and afterwards predicts the reflectance map by a learning-based sparse data interpolation. In order to analyze performance on this difficult task, we propose a new challenge of Specular MAterials on SHapes with complex IllumiNation (SMASHINg) using both synthetic and real images. Furthermore, we show the application of our method to a range of image-based editing tasks on real images.

Rotations performed with the index finger and thumb involve some of the most complex motor action among common multi-touch gestures, yet little is known about the factors affecting performance and ergonomics. This note presents results from a study where the angle, direction, diameter, and position of rotations were systematically manipulated. Subjects were asked to perform the rotations as quickly as possible without losing contact with the display, and were allowed to skip rotations that were too uncomfortable. The data show surprising interaction effects among the variables, and help us identify whole categories of rotations that are slow and cumbersome for users.

Clearly explaining a rationale for a classification decision to an end-user can be as important as the decision itself. Existing approaches for deep visual recognition are generally opaque and do not output any justification text; contemporary vision-language models can describe image content but fail to take into account class-discriminative image aspects which justify visual predictions. We propose a new model that focuses on the discriminating properties of the visible object, jointly predicts a class label, and explains why the predicted label is appropriate for the image. We propose a novel loss function based on sampling and reinforcement learning that learns to generate sentences that realize a global sentence property, such as class specificity. Our results on a fine-grained bird species classification dataset show that our model is able to generate explanations which are not only consistent with an image but also more discriminative than descriptions produced by existing captioning methods.

The goal of this paper is to advance the state-of-the-art of articulated pose estimation in scenes with multiple people. To that end we contribute on three fronts. We propose (1) improved body part detectors that generate effective bottom-up proposals for body parts; (2) novel image-conditioned pairwise terms that allow to assemble the proposals into a variable number of consistent body part configurations; and (3) an incremental optimization strategy that explores the search space more efficiently thus leading both to better performance and significant speed-up factors. We evaluate our approach on two single-person and two multi-person pose estimation benchmarks. The proposed approach significantly outperforms best known multi-person pose estimation results while demonstrating competitive performance on the task of single person pose estimation. Models and code available at http://pose.mpi-inf.mpg.de

With the advent of affordable depth sensors, 3D capture becomes more and more ubiquitous and already has made its way into commercial products. Yet, capturing the geometry or complete shapes of everyday objects using scanning devices (eg. Kinect) still comes with several challenges that result in noise or even incomplete shapes. Recent success in deep learning has shown how to learn complex shape distributions in a data-driven way from large scale 3D CAD Model collections and to utilize them for 3D processing on volumetric representations and thereby circumventing problems of topology and tessellation. Prior work has shown encouraging results on problems ranging from shape completion to recognition. We provide an analysis of such approaches and discover that training as well as the resulting representation are strongly and unnecessarily tied to the notion of object labels. Furthermore, deep learning research argues ~\cite{Vincent08} that learning representation with over-complete model are more prone to overfitting compared to the approach that learns from noisy data. Thus, we investigate a full convolutional volumetric denoising auto encoder that is trained in a unsupervised fashion. It outperforms prior work on recognition as well as more challenging tasks like denoising and shape completion. In addition, our approach is atleast two order of magnitude faster at test time and thus, provides a path to scaling up 3D deep learning.

Graph-based video segmentation methods rely on superpixels as starting point. While most previous work has focused on the construction of the graph edges and weights as well as solving the graph partitioning problem, this paper focuses on better superpixels for video segmentation. We demonstrate by a comparative analysis that superpixels extracted from boundaries perform best, and show that boundary estimation can be significantly improved via image and time domain cues. With superpixels generated from our better boundaries we observe consistent improvement for two video segmentation methods in two different datasets.

We are addressing an open-ended question answering task about real-world images. With the help of currently available methods developed in Computer Vision and Natural Language Processing, we would like to push an architecture with a global visual representation to its limits. In our contribution, we show how to achieve competitive performance on VQA with global visual features (Residual Net) together with a carefully desgined architecture.

Non-maximum suppression (NMS) is used in virtually all state-of-the-art object detection pipelines. While essential object detection ingredients such as features, classifiers, and proposal methods have been extensively researched surprisingly little work has aimed to systematically address NMS. The de-facto standard for NMS is based on greedy clustering with a fixed distance threshold, which forces to trade-off recall versus precision. We propose a convnet designed to perform NMS of a given set of detections. We report experiments on a synthetic setup, and results on crowded pedestrian detection scenes. Our approach overcomes the intrinsic limitations of greedy NMS, obtaining better recall and precision.

Boundary prediction in images as well as video has been a very active topic of research and organizing visual information into boundaries and segments is believed to be a corner stone of visual perception. While prior work has focused on predicting boundaries for observed frames, our work aims at predicting boundaries of future unobserved frames. This requires our model to learn about the fate of boundaries and extrapolate motion patterns. We experiment on established real-world video segmentation dataset, which provides a testbed for this new task. We show for the first time spatio-temporal boundary extrapolation in this challenging scenario. Furthermore, we show long-term prediction of boundaries in situations where the motion is governed by the laws of physics. We successfully predict boundaries in a billiard scenario without any assumptions of a strong parametric model or any object notion. We argue that our model has with minimalistic model assumptions derived a notion of 'intuitive physics' that can be applied to novel scenes.

In this paper we are extracting surface reflectance and natural environmental illumination from a reflectance map, i.e. from a single 2D image of a sphere of one material under one illumination. This is a notoriously difficult problem, yet key to various re-rendering applications. With the recent advances in estimating reflectance maps from 2D images their further decomposition has become increasingly relevant. To this end, we propose a Convolutional Neural Network (CNN) architecture to reconstruct both material parameters (i.e. Phong) as well as illumination (i.e. high-resolution spherical illumination maps), that is solely trained on synthetic data. We demonstrate that decomposition of synthetic as well as real photographs of reflectance maps, both in High Dynamic Range (HDR), and, for the first time, on Low Dynamic Range (LDR) as well. Results are compared to previous approaches quantitatively as well as qualitatively in terms of re-renderings where illumination, material, view or shape are changed.

Recovering natural illumination from a single Low-Dynamic Range (LDR) image is a challenging task. To remedy this situation we exploit two properties often found in everyday images. First, images rarely show a single material, but rather multiple ones that all reflect the same illumination. However, the appearance of each material is observed only for some surface orientations, not all. Second, parts of the illumination are often directly observed in the background, without being affected by reflection. Typically, this directly observed part of the illumination is even smaller. We propose a deep Convolutional Neural Network (CNN) that combines prior knowledge about the statistics of illumination and reflectance with an input that makes explicit use of these two observations. Our approach maps multiple partial LDR material observations represented as reflectance maps and a background image to a spherical High-Dynamic Range (HDR) illumination map. For training and testing we propose a new data set comprising of synthetic and real images with multiple materials observed under the same illumination. Qualitative and quantitative evidence shows how both multi-material and using a background are essential to improve illumination estimations.

Beyond the success in classification, neural networks have recently shown strong results on pixel-wise prediction tasks like image semantic segmentation on RGBD data. However, the commonly used deconvolutional layers for upsampling intermediate representations to the full-resolution output still show different failure modes, like imprecise segmentation boundaries and label mistakes in particular on large, weakly textured objects (e.g. fridge, whiteboard, door). We attribute these errors in part to the rigid way, current network aggregate information, that can be either too local (missing context) or too global (inaccurate boundaries). Therefore we propose a data-driven pooling layer that integrates with fully convolutional architectures and utilizes boundary detection from RGBD image segmentation approaches. We extend our approach to leverage region-level correspondences across images with an additional temporal pooling stage. We evaluate our approach on the NYU-Depth-V2 dataset comprised of indoor RGBD video sequences and compare it to various state-of-the-art baselines. Besides a general improvement over the state-of-the-art, our approach shows particularly good results in terms of accuracy of the predicted boundaries and in segmenting previously problematic classes.

Common computational methods for automated eye movement detection - i.e. the task of detecting different types of eye movement in a continuous stream of gaze data - are limited in that they either involve thresholding on hand-crafted signal features, require individual detectors each only detecting a single movement, or require pre-segmented data. We propose a novel approach for eye movement detection that only involves learning a single detector end-to-end, i.e. directly from the continuous gaze data stream and simultaneously for different eye movements without any manual feature crafting or segmentation. Our method is based on convolutional neural networks (CNN) that recently demonstrated superior performance in a variety of tasks in computer vision, signal processing, and machine learning. We further introduce a novel multi-participant dataset that contains scripted and free-viewing sequences of ground-truth annotated saccades, fixations, and smooth pursuits. We show that our CNN-based method outperforms state-of-the-art baselines by a large margin on this challenging dataset, thereby underlining the significant potential of this approach for holistic, robust, and accurate eye movement protocol analysis.

Recently, Minimum Cost Multicut Formulations have been proposed and proven to be successful in both motion trajectory segmentation and multi-target tracking scenarios. Both tasks benefit from decomposing a graphical model into an optimal number of connected components based on attractive and repulsive pairwise terms. The two tasks are formulated on different levels of granularity and, accordingly, leverage mostly local information for motion segmentation and mostly high-level information for multi-target tracking. In this paper we argue that point trajectories and their local relationships can contribute to the high-level task of multi-target tracking and also argue that high-level cues from object detection and tracking are helpful to solve motion segmentation. We propose a joint graphical model for point trajectories and object detections whose Multicuts are solutions to motion segmentation {\it and} multi-target tracking problems at once. Results on the FBMS59 motion segmentation benchmark as well as on pedestrian tracking sequences from the 2D MOT 2015 benchmark demonstrate the promise of this joint approach.

Understanding physical phenomena is a key competence that enables humans and animals to act and interact under uncertain perception in previously unseen environments containing novel object and their configurations. Developmental psychology has shown that such skills are acquired by infants from observations at a very early stage. In this paper, we contrast a more traditional approach of taking a model-based route with explicit 3D representations and physical simulation by an end-to-end approach that directly predicts stability and related quantities from appearance. We ask the question if and to what extent and quality such a skill can directly be acquired in a data-driven way bypassing the need for an explicit simulation. We present a learning-based approach based on simulated data that predicts stability of towers comprised of wooden blocks under different conditions and quantities related to the potential fall of the towers. The evaluation is carried out on synthetic data and compared to human judgments on the same stimuli.

Together with the development of more accurate methods in Computer Vision and Natural Language Understanding, holistic architectures that answer on questions about the content of real-world images have emerged. In this tutorial, we build a neural-based approach to answer questions about images. We base our tutorial on two datasets: (mostly on) DAQUAR, and (a bit on) VQA. With small tweaks the models that we present here can achieve a competitive performance on both datasets, in fact, they are among the best methods that use a combination of LSTM with a global, full frame CNN representation of an image. We hope that after reading this tutorial, the reader will be able to use Deep Learning frameworks, such as Keras and introduced Kraino, to build various architectures that will lead to a further performance improvement on this challenging task.

Deep models are the defacto standard in visual decision models due to their impressive performance on a wide array of visual tasks. However, they are frequently seen as opaque and are unable to explain their decisions. In contrast, humans can justify their decisions with natural language and point to the evidence in the visual world which led to their decisions. We postulate that deep models can do this as well and propose our Pointing and Justification (PJ-X) model which can justify its decision with a sentence and point to the evidence by introspecting its decision and explanation process using an attention mechanism. Unfortunately there is no dataset available with reference explanations for visual decision making. We thus collect two datasets in two domains where it is interesting and challenging to explain decisions. First, we extend the visual question answering task to not only provide an answer but also a natural language explanation for the answer. Second, we focus on explaining human activities which is traditionally more challenging than object classification. We extensively evaluate our PJ-X model, both on the justification and pointing tasks, by comparing it to prior models and ablations using both automatic and human evaluations.

Marker-based and marker-less optical skeletal motion-capture methods use an outside-in arrangement of cameras placed around a scene, with viewpoints converging on the center. They often create discomfort by possibly needed marker suits, and their recording volume is severely restricted and often constrained to indoor scenes with controlled backgrounds. We therefore propose a new method for real-time, marker-less and egocentric motion capture which estimates the full-body skeleton pose from a lightweight stereo pair of fisheye cameras that are attached to a helmet or virtual-reality headset. It combines the strength of a new generative pose estimation framework for fisheye views with a ConvNet-based body-part detector trained on a new automatically annotated and augmented dataset. Our inside-in method captures full-body motion in general indoor and outdoor scenes, and also crowded scenes.

Gaze reflects how humans process visual scenes and is therefore increasingly used in computer vision systems. Previous works demonstrated the potential of gaze for object-centric tasks, such as object localization and recognition, but it remains unclear if gaze can also be beneficial for scene-centric tasks, such as image captioning. We present a new perspective on gaze-assisted image captioning by studying the interplay between human gaze and the attention mechanism of deep neural networks. Using a public large-scale gaze dataset, we first assess the relationship between state-of-the-art object and scene recognition models, bottom-up visual saliency, and human gaze. We then propose a novel split attention model for image captioning. Our model integrates human gaze information into an attention-based long short-term memory architecture, and allows the algorithm to allocate attention selectively to both fixated and non-fixated image regions. Through evaluation on the COCO/SALICON datasets we show that our method improves image captioning performance and that gaze can complement machine attention for semantic scene understanding tasks.

The eyes are a rich channel for non-verbal communication in our daily interactions. We propose social gaze interaction as a game mechanic to enhance user interactions with virtual characters. We develop a game from the ground-up in which characters are esigned to be reactive to the player’s gaze in social ways, such as etting annoyed when the player seems distracted or changing their dialogue depending on the player’s apparent focus of ttention. Results from a qualitative user study provide insights bout how social gaze interaction is intuitive for users, elicits deep feelings of immersion, and highlight the players’ self-consciousness of their own eye movements through their strong reactions to the characters

Head-mounted eye tracking has significant potential for mobile gaze-based interaction with ambient displays but current interfaces lack information about the tracker\'s gaze estimation error. Consequently, current interfaces do not exploit the full potential of gaze input as the inherent estimation error can not be dealt with. The error depends on the physical properties of the display and constantly varies with changes in position and distance of the user to the display. In this work we present a computational model of gaze estimation error for head-mounted eye trackers. Our model covers the full processing pipeline for mobile gaze estimation, namely mapping of pupil positions to scene camera coordinates, marker-based display detection, and display mapping. We build the model based on a series of controlled measurements of a sample state-of-the-art monocular head-mounted eye tracker. Results show that our model can predict gaze estimation error with a root mean squared error of 17.99~px ($1.96^\\circ$).

Mobile gaze-based interaction with multiple displays may occur from arbitrary positions and orientations. However, maintaining high gaze estimation accuracy still represents a significant challenge. To address this, we present GazeProjector, a system that combines accurate point-of-gaze estimation with natural feature tracking on displays to determine the mobile eye tracker’s position relative to a display. The detected eye positions are transformed onto that display allowing for gaze-based interaction. This allows for seamless gaze estimation and interaction on (1) multiple displays of arbitrary sizes, (2) independently of the user’s position and orientation to the display. In a user study with 12 participants we compared GazeProjector to existing well- established methods such as visual on-screen markers and a state-of-the-art motion capture system. Our results show that our approach is robust to varying head poses, orientations, and distances to the display, while still providing high gaze estimation accuracy across multiple displays without re-calibration. The system represents an important step towards the vision of pervasive gaze-based interfaces.

Szeliski et al. published an influential study in 2006 on energy minimization methods for Markov Random Fields (MRF). This study provided valuable insights in choosing the best optimization technique for certain classes of problems. While these insights remain generally useful today, the phenomenal success of random field models means that the kinds of inference problems that have to be solved changed significantly. Specifically, the models today often include higher order interactions, flexible connectivity structures, large la\-bel-spaces of different cardinalities, or learned energy tables. To reflect these changes, we provide a modernized and enlarged study. We present an empirical comparison of 32 state-of-the-art optimization techniques on a corpus of 2,453 energy minimization instances from diverse applications in computer vision. To ensure reproducibility, we evaluate all methods in the OpenGM 2 framework and report extensive results regarding runtime and solution quality. Key insights from our study agree with the results of Szeliski et al. for the types of models they studied. However, on new and challenging types of models our findings disagree and suggest that polyhedral methods and integer programming solvers are competitive in terms of runtime and solution quality over a large range of model types.

Progress in language and image understanding by machines has sparkled the interest of the research community in more open-ended, holistic tasks, and refueled an old AI dream of building intelligent machines. We discuss a few prominent challenges that characterize such holistic tasks and argue for "question answering about images" as a particular appealing instance of such a holistic task. In particular, we point out that it is a version of a Turing Test that is likely to be more robust to over-interpretations and contrast it with tasks like grounding and generation of descriptions. Finally, we discuss tools to measure progress in this field.

Current top performing object detectors employ detection proposals to guide the search for objects, thereby avoiding exhaustive sliding window search across images. Despite the popularity and widespread use of detection proposals, it is unclear which trade-offs are made when using them during object detection. We provide an in-depth analysis of twelve proposal methods along with four baselines regarding proposal repeatability, ground truth annotation recall on PASCAL and ImageNet, and impact on DPM and R-CNN detection performance. Our analysis shows that for object detection improving proposal localisation accuracy is as important as improving recall. We introduce a novel metric, the average recall (AR), which rewards both high recall and good localisation and correlates surprisingly well with detector performance. Our findings show common strengths and weaknesses of existing methods, and provide insights and metrics for selecting and tuning proposal methods.