2017
STD2P: RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling
Y. He, W.-C. Chiu, M. Keuper and M. Fritz
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Learning Non-maximum Suppression
J. Hosang, R. Benenson and B. Schiele
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Gaze Embeddings for Zero-Shot Image Classification
N. Karessli, Z. Akata, B. Schiele and A. Bulling
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Simple Does It: Weakly Supervised Instance and Semantic Segmentation
A. Khoreva, R. Benenson, J. Hosang, M. Hein and B. Schiele
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Learning Video Object Segmentation from Static Images
A. Khoreva, F. Perazzi, R. Benenson, B. Schiele and A. Sorkine-Hornung
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Joint Graph Decomposition and Node Labeling: Problem, Algorithms, Applications
E. Levinkov, J. Uhrig, S. Tang, M. Omran, E. Insafutdinov, A. Kirillov, C. Rother, T. Brox, B. Schiele and B. Andres
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
A Dataset and Exploration of Models for Understanding Video Data through Fill-in-the-blank Question-answering
T. Maharaj, N. Ballas, A. Rohrbach, A. Courville and C. Pal
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Exploiting Saliency for Object Segmentation from Image Level Labels
S. J. Oh, R. Benenson, A. Khoreva, Z. Akata, M. Fritz and B. Schiele
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Generating Descriptions with Grounded and Co-Referenced People
A. Rohrbach, M. Rohrbach, S. Tang, S. J. Oh and B. Schiele
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
A Domain Based Approach to Social Relation Recognition
Q. Sun, B. Schiele and M. Fritz
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Zero-shot learning - The Good, the Bad and the Ugly
Y. Xian, B. Schiele and Z. Akata
30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), 2017
(Accepted/in press)
Noticeable or Distractive? A Design Space for Gaze-Contingent User Interface Notifications
M. Klauck, Y. Sugano and A. Bulling
CHI 2017 Extended Abstracts, 2017
Visual Stability Prediction for Robotic Manipulation
W. Li, A. Leonardis and M. Fritz
IEEE International Conference on Robotics and Automation (ICRA 2017), 2017
(Accepted/in press)
MARCOnI-ConvNet-Based MARker-Less Motion Capture in Outdoor and Indoor Scenes
A. Elhayek, E. de Aguiar, A. Jain, J. Thompson, L. Pishchulin, M. Andriluka, C. Bregler, B. Schiele and C. Theobalt
IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume 33, Number 3, 2017
Novel Views of Objects from a Single Image
K. Rematas, C. Nguyen, T. Ritschel, M. Fritz and T. Tuytelaars
IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume 39, Number 8, 2017
Expanded Parts Model for Semantic Description of Humans in Still Images
G. Sharma, F. Jurie and C. Schmid
IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume 39, Number 1, 2017
Towards Reaching Human Performance in Pedestrian Detection
S. Zhang, R. Benenson, M. Omran, J. Hosang and B. Schiele
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017
Abstract
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.
Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Images
T. Orekondy, B. Schiele and M. Fritz
International Conference on Computer Vision (ICCV 2017), 2017
(Accepted/in press)
Movie Description
A. Rohrbach, A. Torabi, M. Rohrbach, N. Tandon, C. Pal, H. Larochelle, A. Courville and B. Schiele
International Journal of Computer Vision, Volume 123, Number 1, 2017
Abstract
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.
Building Statistical Shape Spaces for 3D Human Modeling
L. Pishchulin, S. Wuhrer, T. Helten, C. Theobalt and B. Schiele
Pattern Recognition, Volume 67, 2017
Online Growing Neural Gas for Anomaly Detection in Changing Surveillance Scenes
Q. Sun, H. Liu and T. Harada
Pattern Recognition, Volume 64, 2017
Learning Dilation Factors for Semantic Segmentation of Street Scenes
Y. He, M. Keuper, B. Schiele and M. Fritz
Pattern Recognition (GCPR 2017), 2017
(Accepted/in press)
Look Together: Using Gaze for Assisting Co-located Collaborative Search
Y. Zhang, K. Pfeuffer, M. K. Chong, J. Alexander, A. Bulling and H. Gellersen
Personal and Ubiquitous Computing, Volume 21, Number 1, 2017
Efficiently Summarising Event Sequences with Rich Interleaving Patterns
A. Bhattacharyya and J. Vreeken
Proceedings of the Seventeenth SIAM International Conference on Data Mining (SDM 2017), 2017
(Accepted/in press)
Advanced Steel Microstructure Classification by Deep Learning Methods
S. M. Azimi, D. Britz, M. Engstler, M. Fritz and F. Mücklich
Technical Report, 2017
(arXiv: 1706.06480)
Abstract
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.
Analysis and Improvement of the Visual Object Detection Pipeline
J. Hosang
PhD Thesis, Universität des Saarlandes, 2017
Abstract
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.
Lucid Data Dreaming for Object Tracking
A. Khoreva, R. Benenson, E. Ilg, T. Brox and B. Schiele
Technical Report, 2017
(arXiv: 1703.09554)
Abstract
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.
Decomposition of Trees and Paths via Correlation
J.-H. Lange and B. Andres
Technical Report, 2017
(arXiv: 1706.06822)
Abstract
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, we show (i) the objective function is not necessarily submodular, (ii) the problem can be solved in quasi-polynomial time by divide-and-conquer, (iii) the problem can be solved in polynomial time by polyhedral optimization techniques. To establish (iii), we offer a totally dual integral (TDI) description of path partition polytopes.
Image Classification with Limited Training Data and Class Ambiguity
M. Lapin
PhD Thesis, Universität des Saarlandes, 2017
Abstract
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.
Discrete-Continuous Splitting for Weakly Supervised Learning
E. Laude, J.-H. Lange, F. R. Schmidt, B. Andres and D. Cremers
Technical Report, 2017
(arXiv: 1705.05020)
Abstract
This paper introduces a novel algorithm for a class of weakly supervised learning tasks. The considered tasks are posed as joint optimization problems in the continuous model parameters and the (a-priori unknown) discrete label variables. In contrast to prior approaches such as convex relaxations, we decompose the nonconvex problem into purely discrete and purely continuous subproblems in a way that is amenable to distributed optimization by the Alternating Direction Method of Multipliers (ADMM). This approach preserves integrality of the discrete label variables and, for a reparameterized variant of the algorithm using kernels, guarantees global convergence to a critical point. The resulting method implicitly alternates between a discrete and a continuous variable update, however, it is inherently different from a discrete-continuous coordinate descent scheme (hard EM). In diverse experiments we show that our method can learn a classifier from weak supervision that takes the form of hard and soft constraints on the labeling and outperforms hard EM in this task.
Towards Holistic Machines: From Visual Recognition To Question Answering About Real-world Image
M. Malinowski
PhD Thesis, Universität des Saarlandes, 2017
Abstract
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.
Pose Guided Person Image Generation
L. Ma, X. Jia, Q. Sun, B. Schiele, T. Tuytelaars and L. Van Gool
Technical Report, 2017
(arXiv: 1705.09368)
Abstract
This paper proposes the novel Pose Guided Person Generation Network (PG$^2$) that allows to synthesize person images in arbitrary poses, based on an image of that person and a novel pose. Our generation framework PG$^2$ utilizes the pose information explicitly and consists of two key stages: pose integration and image refinement. In the first stage the condition image and the target pose are fed into a U-Net-like network to generate an initial but coarse image of the person with the target pose. The second stage then refines the initial and blurry result by training a U-Net-like generator in an adversarial way. Extensive experimental results on both 128$\times$64 re-identification images and 256$\times$256 fashion photos show that our model generates high-quality person images with convincing details.
Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspective
S. J. Oh, M. Fritz and B. Schiele
Technical Report, 2017
(arXiv: 1703.09471)
Abstract
Users like sharing personal photos with others through social media. At the same time, they might want to make automatic identification in such photos difficult or even impossible. Classic obfuscation methods such as blurring are not only unpleasant but also not as effective as one would expect. Recent studies on adversarial image perturbations (AIP) suggest that it is possible to confuse recognition systems effectively without unpleasant artifacts. However, in the presence of counter measures against AIPs, it is unclear how effective AIP would be in particular when the choice of counter measure is unknown. Game theory provides tools for studying the interaction between agents with uncertainties in the strategies. We introduce a general game theoretical framework for the user-recogniser dynamics, and present a case study that involves current state of the art AIP and person recognition techniques. We derive the optimal strategy for the user that assures an upper bound on the recognition rate independent of the recogniser's counter measure.
Efficient Algorithms for Moral Lineage Tracing
M. Rempfler, J.-H. Lange, F. Jug, C. Blasse, E. W. Myers, B. H. Menze and B. Andres
Technical Report, 2017
(arXiv: 1702.04111)
Abstract
Lineage tracing, the joint segmentation and tracking of living cells as they move and divide in a sequence of light microscopy images, is a challenging task. Jug et al. have proposed a mathematical abstraction of this task, the moral lineage tracing problem (MLTP) whose feasible solutions define a segmentation of every image and a lineage forest of cells. Their branch-and-cut algorithm, however, is prone to many cuts and slow convergences for large instances. To address this problem, we make three contributions: Firstly, we improve the branch-and-cut algorithm by separating tighter cutting planes. Secondly, we define two primal feasible local search algorithms for the MLTP. Thirdly, we show in experiments that our algorithms decrease the runtime on the problem instances of Jug et al. considerably and find solutions on larger instances in reasonable time.
Generation and Grounding of Natural Language Descriptions for Visual Data
A. Rohrbach
PhD Thesis, universität des Saarlandes, 2017
Abstract
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.
Visual Decoding of Targets During Visual Search From Human Eye Fixations
H. Sattar, M. Fritz and A. Bulling
Technical Report, 2017
(arXiv: 1706.05993)
Abstract
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
Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training
R. Shetty, M. Rohrbach, L. A. Hendricks, M. Fritz and B. Schiele
Technical Report, 2017
(arXiv: 1703.10476)
Abstract
While strong progress has been made in image captioning over the last years, machine and human captions are still quite distinct. A closer look reveals that this is due to the deficiencies in the generated word distribution, vocabulary size, and strong bias in the generators towards frequent captions. Furthermore, humans -- rightfully so -- generate multiple, diverse captions, due to the inherent ambiguity in the captioning task which is not considered in today's systems. To address these challenges, we change the training objective of the caption generator from reproducing groundtruth captions to generating a set of captions that is indistinguishable from human generated captions. Instead of handcrafting such a learning target, we employ adversarial training in combination with an approximate Gumbel sampler to implicitly match the generated distribution to the human one. While our method achieves comparable performance to the state-of-the-art in terms of the correctness of the captions, we generate a set of diverse captions, that are significantly less biased and match the word statistics better in several aspects.
GazeDirector: Fully Articulated Eye Gaze Redirection in Video
E. Wood, T. Baltrusaitis, L.-P. Morency, P. Robinson and A. Bulling
Technical Report, 2017
(arXiv: 1704.08763)
Abstract
We present GazeDirector, a new approach for eye gaze redirection that uses model-fitting. Our method first tracks the eyes by fitting a multi-part eye region model to video frames using analysis-by-synthesis, thereby recovering eye region shape, texture, pose, and gaze simultaneously. It then redirects gaze by 1) warping the eyelids from the original image using a model-derived flow field, and 2) rendering and compositing synthesized 3D eyeballs onto the output image in a photorealistic manner. GazeDirector allows us to change where people are looking without person-specific training data, and with full articulation, i.e. we can precisely specify new gaze directions in 3D. Quantitatively, we evaluate both model-fitting and gaze synthesis, with experiments for gaze estimation and redirection on the Columbia gaze dataset. Qualitatively, we compare GazeDirector against recent work on gaze redirection, showing better results especially for large redirection angles. Finally, we demonstrate gaze redirection on YouTube videos by introducing new 3D gaze targets and by manipulating visual behavior.
CityPersons: A Diverse Dataset for Pedestrian Detection
S. Zhang, R. Benenson and B. Schiele
Technical Report, 2017
(arXiv: 1702.05693)
Abstract
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.