Saurabh Sharma

Personal Information

Researcher at the Computer Vision and Machine Learning group in the Max Planck Institute for Informatics.
Masters student in Computer Science at Saarland University.
[Linkedin] [Github] [Google Scholar]

Publications

2020

S. Sharma, N. Yu, M. Fritz, and B. Schiele

“Long-Tailed Recognition Using Class-Balanced Experts,” in Pattern Recognition (GCPR 2020), Tübingen, Germany, 2021.

@inproceedings{Sharma_GCPR2020,
TITLE = {Long-Tailed Recognition Using Class-Balanced Experts},
AUTHOR = {Sharma, Saurabh and Yu, Ning and Fritz, Mario and Schiele, Bernt},
LANGUAGE = {eng},
ISBN = {978-3-030-71277-8},
DOI = {10.1007/978-3-030-71278-5_7},
PUBLISHER = {Springer},
YEAR = {2020},
DATE = {2021},
BOOKTITLE = {Pattern Recognition (GCPR 2020)},
EDITOR = {Akata, Zeynep and Geiger, Andreas and Sattler, Torsten},
PAGES = {86--100},
SERIES = {Lecture Notes in Computer Science},
ADDRESS = {T{\"u}bingen, Germany},
}

Endnote

%0 Conference Proceedings
%A Sharma, Saurabh
%A Yu, Ning
%A Fritz, Mario
%A Schiele, Bernt
%+ Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
External Organizations
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
%T Long-Tailed Recognition Using Class-Balanced Experts : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0007-80CE-0
%R 10.1007/978-3-030-71278-5_7
%D 2021
%B 42nd German Conference on Pattern Recognition
%Z date of event: 2020-09-28 - 2020-10-01
%C T&#252;bingen, Germany
%B Pattern Recognition
%E Akata, Zeynep; Geiger, Andreas; Sattler, Torsten
%P 86 - 100
%I Springer
%@ 978-3-030-71277-8
%B Lecture Notes in Computer Science

2019

Y. Xian, S. Sharma, B. Schiele, and Z. Akata

“f-VAEGAN-D2: A Feature Generating Framework for Any-Shot Learning,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2019), Long Beach, CA, USA, 2019.

Abstract

When labeled training data is scarce, a promising data augmentation approach is to generate visual features of unknown classes using their attributes. To learn the class conditional distribution of CNN features, these models rely on pairs of image features and class attributes. Hence, they can not make use of the abundance of unlabeled data samples. In this paper, we tackle any-shot learning problems i.e. zero-shot and few-shot, in a unified feature generating framework that operates in both inductive and transductive learning settings. We develop a conditional generative model that combines the strength of VAE and GANs and in addition, via an unconditional discriminator, learns the marginal feature distribution of unlabeled images. We empirically show that our model learns highly discriminative CNN features for five datasets, i.e. CUB, SUN, AWA and ImageNet, and establish a new state-of-the-art in any-shot learning, i.e. inductive and transductive (generalized) zero- and few-shot learning settings. We also demonstrate that our learned features are interpretable: we visualize them by inverting them back to the pixel space and we explain them by generating textual arguments of why they are associated with a certain label.

BibTeX

@inproceedings{XianCVPR2019a,
TITLE = {{\texttt{f-VAEGAN-D2}}: {A} Feature Generating Framework for Any-Shot Learning},
AUTHOR = {Xian, Yongqin and Sharma, Saurabh and Schiele, Bernt and Akata, Zeynep},
LANGUAGE = {eng},
ISBN = {978-1-7281-3293-8},
DOI = {10.1109/CVPR.2019.01052},
PUBLISHER = {IEEE},
YEAR = {2019},
ABSTRACT = {When labeled training data is scarce, a promising data augmentation approach is to generate visual features of unknown classes using their attributes. To learn the class conditional distribution of CNN features, these models rely on pairs of image features and class attributes. Hence, they can not make use of the abundance of unlabeled data samples. In this paper, we tackle any-shot learning problems i.e. zero-shot and few-shot, in a unified feature generating framework that operates in both inductive and transductive learning settings. We develop a conditional generative model that combines the strength of VAE and GANs and in addition, via an unconditional discriminator, learns the marginal feature distribution of unlabeled images. We empirically show that our model learns highly discriminative CNN features for five datasets, i.e. CUB, SUN, AWA and ImageNet, and establish a new state-of-the-art in any-shot learning, i.e. inductive and transductive (generalized) zero- and few-shot learning settings. We also demonstrate that our learned features are interpretable: we visualize them by inverting them back to the pixel space and we explain them by generating textual arguments of why they are associated with a certain label.},
BOOKTITLE = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2019)},
PAGES = {10275--10284},
ADDRESS = {Long Beach, CA, USA},
}

Endnote

%0 Conference Proceedings
%A Xian, Yongqin
%A Sharma, Saurabh
%A Schiele, Bernt
%A Akata, Zeynep
%+ Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
%T f-VAEGAN-D2: A Feature Generating Framework for Any-Shot 
Learning : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0003-162E-2
%R 10.1109/CVPR.2019.01052
%D 2019
%B 32nd IEEE Conference on Computer Vision and Pattern Recognition
%Z date of event: 2019-06-16 - 2019-06-20
%C Long Beach, CA,  USA
%X When labeled training data is scarce, a promising data augmentation approach is to generate visual features of unknown classes using their attributes. To learn the class conditional distribution of CNN features, these models rely on pairs of image features and class attributes. Hence, they can not make use of the abundance of unlabeled data samples. In this paper, we tackle any-shot learning problems i.e. zero-shot and few-shot, in a unified feature generating framework that operates in both inductive and transductive learning settings. We develop a conditional generative model that combines the strength of VAE and GANs and in addition, via an unconditional discriminator, learns the marginal feature distribution of unlabeled images. We empirically show that our model learns highly discriminative CNN features for five datasets, i.e. CUB, SUN, AWA and ImageNet, and establish a new state-of-the-art in any-shot learning, i.e. inductive and transductive (generalized) zero- and few-shot learning settings. We also demonstrate that our learned features are interpretable: we visualize them by inverting them back to the pixel space and we explain them by generating textual arguments of why they are associated with a certain label.
%B IEEE/CVF Conference on Computer Vision and Pattern Recognition
%P 10275 - 10284
%I IEEE
%@ 978-1-7281-3293-8
%U http://openaccess.thecvf.com/content_CVPR_2019/html/Xian_F-VAEGAN-D2_A_Feature_Generating_Framework_for_Any-Shot_Learning_CVPR_2019_paper.html

S. Sharma, P. T. Varigonda, P. Bindal, A. Sharma, and A. Jain

“Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking,” in International Conference on Computer Vision (ICCV 2019), Seoul, Korea, 2019.

@inproceedings{sharma2019_ICCV2019,
TITLE = {Monocular 3{D} Human Pose Estimation by Generation and Ordinal Ranking},
AUTHOR = {Sharma, Saurabh and Varigonda, Pavan Teja and Bindal, Prashast and Sharma, Abhishek and Jain, Arjun},
LANGUAGE = {eng},
ISBN = {978-1-7281-4803-8},
DOI = {10.1109/ICCV.2019.00241},
PUBLISHER = {IEEE},
YEAR = {2019},
DATE = {2019},
BOOKTITLE = {International Conference on Computer Vision (ICCV 2019)},
PAGES = {2325--2334},
ADDRESS = {Seoul, Korea},
}

Endnote

%0 Conference Proceedings
%A Sharma, Saurabh
%A Varigonda, Pavan Teja
%A Bindal, Prashast
%A Sharma, Abhishek
%A Jain, Arjun
%+ Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society
External Organizations
External Organizations
External Organizations
External Organizations
%T Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking : 
%G eng
%U http://hdl.handle.net/21.11116/0000-0004-59A8-B
%R 10.1109/ICCV.2019.00241
%D 2019
%B International Conference on Computer Vision
%Z date of event: 2019-10-27 - 2019-11-02
%C Seoul, Korea
%B International Conference on Computer Vision
%P 2325 - 2334
%I IEEE
%@ 978-1-7281-4803-8
%U https://github.com/ssfootball04/generative_pose