Subhabrata Choudhury
Education
Saarland University
MSc, Computer Science
Germany, 2017-2019
IIT Kharagpur
BTech (Hons), Metallurgical and Materials Engineering
India, 2009-2013
Experience
Publications
Unsupervised Multi-Object Segmentation by Predicting Probable Motion Patterns
NeurIPS 2022
pdf ⋄
project page ⋄
code ⋄
arXiv ⋄
BibTeX
@inproceedings{karazija+choudhury2022unsupervised,
author = {Karazija, Laurynas and Choudhury, Subhabrata and Laina, Iro and Rupprecht, Christian and Vedaldi, Andrea},
booktitle = {Advances in Neural Information Processing Systems},
title = {Unsupervised Multi-Object Segmentation by Predicting Probable Motion Patterns},
year = {2022},
}
We learn to segment independent objects in still images by predicting regions that contain motion patterns likely to arise from such objects.
Guess What Moves: Unsupervised Video and Image Segmentation by Anticipating Motion
BMVC 2022 (Spotlight)
pdf ⋄
project page ⋄
video ⋄
code ⋄
arXiv ⋄
BibTeX
@inproceedings{choudhury+karazija2022guess,
author = {Choudhury, Subhabrata and Karazija, Laurynas and Laina, Iro and Vedaldi, Andrea and Rupprecht, Christian},
booktitle = {British Machine Vision Conference}
title = {Guess What Moves: Unsupervised Video and Image Segmentation by Anticipating Motion}
year = {2022}
}
We use motion anticipation as a learning signal to train an image segmentation network to predict regions that likely contain similar optical flow patterns
Unsupervised Part Discovery from Contrastive Reconstruction
NeurIPS 2021
pdf ⋄
project page ⋄
video ⋄
code ⋄
arXiv ⋄
BibTeX
@inproceedings{choudhury2021unsupervised,
author = {Choudhury, Subhabrata and Laina, Iro and Rupprecht, Christian and Vedaldi, Andrea},
booktitle = {Advances in Neural Information Processing Systems},
title = {Unsupervised Part Discovery from Contrastive Reconstruction},
volume = {35},
year = {2021},
}
Unsupervised discovery of semantically meaningful parts of objects using appearance consistency cues and contrastive learning
The Curious Layperson: Fine-Grained Image Recognition without Expert Labels
BMVC 2021 (Oral) [ Best Student Paper Award ]
pdf ⋄
project page ⋄
video ⋄
code ⋄
arXiv ⋄
BibTeX
@inproceedings{choudhury2021curious,
author = {Choudhury, Subhabrata and Laina, Iro and Rupprecht, Christian and Vedaldi, Andrea},
booktitle = {British Machine Vision Conference},
title = {The Curious Layperson: Fine-Grained Image Recognition without Expert Labels},
volume = {32},
year = {2021},
}
Proposes a new problem of fine-grained image classification without expert annotations, by utilizing class agnostic non-expert descriptions and off-the-shelf expert corpus
Semantic Projection Network for Zero- and Few-Label Semantic Segmentation
CVPR 2019
pdf ⋄
project page ⋄
code ⋄
BibTeX
@InProceedings{xian+choudhury2019semantic,
author = {Xian, Yongqin and Choudhury, Subhabrata and He, Yang and Schiele, Bernt and Akata, Zeynep},
title = {Semantic Projection Network for Zero- and Few-Label Semantic Segmentation},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2019}
}
Extenstion of zero-shot learning to segmentation: perform dense labelling of 'unseen' test classes by training on mutually exclusive 'seen' training classes with additional help from side information