Jean Kossaifi

Dr. Jean Kossaifi

Position:

Past Member

Email:

Personal website:

http://jeankossaifi.com/

Biography

Jean Kossaifi is a Senior Research Scientist at NVIDIA and a Research Associate in the Department of Computing, Imperial College London. Prior to this, he was a Research Scientist at the Samsung AI Center in Cambridge. He has worked extensively on face analysis and facial affect estimation in naturalistic conditions, a field which bridges the gap between computer vision and machine learning. His current focus is tensor methods for machine learning and efficient combination of these methods with deep learning to develop better models that are memory and computation efficient, while being more robust to noise, random or adversarial, as well as domain shift. He is the creator of TensorLy, a high-level API for tensor methods and deep tensorized neural networks in Python, designed to make tensor learning simple and accessible.

Projects:

sewa-project, TERESA

Publications

Journal articles

Disentangling Geometry and Appearance with Regularised Geometry-Aware Generative Adversarial Networks

L. Tran, J. Kossaifi, Y. Panagakis, M. Pantic. International Journal of Computer Vision (IJCV). 2019.

Bibtex reference [hide]
@article{GAGAN-IJCV,
    author = {L. Tran and J. Kossaifi and Y. Panagakis and M. Pantic},
    journal = {International Journal of Computer Vision (IJCV)},
    title = {Disentangling Geometry and Appearance with Regularised Geometry-Aware Generative Adversarial Networks},
    year = {2019},
}
Endnote reference [hide]

TensorLy: Tensor Learning in Python

J. Kossaifi, Y. Panagakis, A. Anandkumar, M. Pantic. Journal of Machine Learning Research (JMLR). 20(26): pp. 1 - 6, 2019.

Bibtex reference [hide]
@article{tensorly,
    author = {J. Kossaifi and Y. Panagakis and A. Anandkumar and M. Pantic},
    pages = {1--6},
    journal = {Journal of Machine Learning Research (JMLR)},
    number = {26},
    title = {TensorLy: Tensor Learning in Python},
    url = {http://jmlr.org/papers/v20/18-277.html},
    volume = {20},
    year = {2019},
}
Endnote reference [hide]

AFEW-VA database for valence and arousal estimation in-the-wild

J. Kossaifi, G. Tzimiropoulos, S. Todorovic, M. Pantic. Image and Vision Computing. 65: pp. 23 - 36, Multimodal Sentiment Analysis and Mining in the Wild Image and Vision Computing. 2017.

Bibtex reference [hide]
@article{kossaifi_afewva,
    author = {J. Kossaifi and G. Tzimiropoulos and S. Todorovic and M. Pantic},
    pages = {23--36},
    journal = {Image and Vision Computing},
    note = {Multimodal Sentiment Analysis and Mining in the Wild Image and Vision Computing},
    title = {AFEW-VA database for valence and arousal estimation in-the-wild},
    volume = {65},
    year = {2017},
}
Endnote reference [hide]

Fast and exact Newton and Bidirectional fitting of Active Appearance Models

J. Kossaifi, G. Tzimiropoulos, M. Pantic. IEEE Transactions on Image Processing (TIP). 26: pp. 1040 - 1053, 2016.

Bibtex reference [hide]
@article{kossaifi_aam_tip,
    author = {J. Kossaifi and G. Tzimiropoulos and M. Pantic},
    pages = {1040--1053},
    journal = {IEEE Transactions on Image Processing (TIP)},
    publisher = {IEEE},
    title = {Fast and exact Newton and Bidirectional fitting of Active Appearance Models},
    volume = {26},
    year = {2016},
}
Endnote reference [hide]

Machine Learning for Neuroimaging with Scikit-Learn

A. Abraham, F. Pedregosa, M. Eickenberg, P. Gervais, A. Mueller, J. Kossaifi, A. Gramfort, B. Thirion, G. Varoquaux. Frontiers in Neuroinformatics. 8(14): 2014.

Bibtex reference [hide]
@article{ml_neuroimaging_sklearn,
    author = {A. Abraham and F. Pedregosa and M. Eickenberg and P. Gervais and A. Mueller and J. Kossaifi and A. Gramfort and B. Thirion and G. Varoquaux},
    journal = {Frontiers in Neuroinformatics},
    number = {14},
    title = {Machine Learning for Neuroimaging with Scikit-Learn},
    url = {http://www.frontiersin.org/neuroinformatics/10.3389/fninf.2014.00014/abstract},
    volume = {8},
    year = {2014},
}
Endnote reference [hide]

Conference articles

Augmenting Deep Classifiers with Polynomial Neural Networks

G. Chrysos, M. Georgopoulos, J. Deng, J. Kossaifi, Y. Panagakis, A. Anandkumar. ECCV. 2022.

Bibtex reference [hide]
@inproceedings{ECCV2022Polynomial,
    booktitle = {ECCV},
    title = {Augmenting Deep Classifiers with Polynomial Neural Networks},
    year = {2022},
}
Endnote reference [hide]

T-Net: Parametrizing Fully Convolutional Nets with a Single High-Order Tensor

J. Kossaifi, A. Bulat, G. Tzimiropoulos, M. Pantic. IEEE CVPR. 2019.

Bibtex reference [hide]
@inproceedings{kossaifi2019tnet,
    author = {J. Kossaifi and A. Bulat and G. Tzimiropoulos and M. Pantic},
    booktitle = {IEEE CVPR},
    title = {T-Net: Parametrizing Fully Convolutional Nets with a Single High-Order Tensor},
    year = {2019},
}
Endnote reference [hide]

Robust Conditional Generative Adversarial Networks

G. Chrysos, J. Kossaifi, S. Zafeiriou. ICLR. May 2019.

Bibtex reference [hide]
@inproceedings{chrysos2019,
    booktitle = {ICLR},
    month = {May},
    title = {Robust Conditional Generative Adversarial Networks},
    year = {2019},
}
Endnote reference [hide]

GAGAN: Geometry-Aware Generative Adversarial Networks

J. Kossaifi, L. Tran, Y. Panagakis, M. Pantic. IEEE CVPR. pp. 878 - 887, June 2018.

Bibtex reference [hide]
@inproceedings{kossaifi2018gagan,
    author = {J. Kossaifi and L. Tran and Y. Panagakis and M. Pantic},
    pages = {878--887},
    booktitle = {IEEE CVPR},
    month = {June},
    title = {GAGAN: Geometry-Aware Generative Adversarial Networks},
    year = {2018},
}
Endnote reference [hide]

Stochastic activation pruning for robust adversarial defense

G. S. Dhillon, K. Azizzadenesheli, Z. Lipton, J. Bernstein, J. Kossaifi, A. Khanna, A. Anandkumar. International Conference on Learning Representations (ICLR). 2018.

Bibtex reference [hide]
@inproceedings{dhillon2018stochastic,
    author = {G. S. Dhillon and K. Azizzadenesheli and Z. Lipton and J. Bernstein and J. Kossaifi and A. Khanna and A. Anandkumar},
    booktitle = {International Conference on Learning Representations (ICLR)},
    title = {Stochastic activation pruning for robust adversarial defense},
    year = {2018},
}
Endnote reference [hide]

Tensor Regression Networks

J. Kossaifi, Z. Lipton, A. Khanna, T. Furlanello, A. Anandkumar. 2017.

Bibtex reference [hide]
@inproceedings{kossaifi_trl,
    author = {J. Kossaifi and Z. Lipton and A. Khanna and T. Furlanello and A. Anandkumar},
    journal = {CoRR},
    title = {Tensor Regression Networks},
    url = {http://arxiv.org/abs/1707.08308},
    year = {2017},
}
Endnote reference [hide]

Tensor Contraction Layers for Parsimonious Deep Nets

J. Kossaifi, A. Khanna, Z. Lipton, T. Furlanello, A. Anandkumar. 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). pp. 1940 - 1946, July 2017.

Bibtex reference [hide]
@inproceedings{kossaifi_tcl,
    author = {J. Kossaifi and A. Khanna and Z. Lipton and T. Furlanello and A. Anandkumar},
    pages = {1940--1946},
    booktitle = {2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
    month = {July},
    title = {Tensor Contraction Layers for Parsimonious Deep Nets},
    year = {2017},
}
Endnote reference [hide]

The First Facial Landmark Tracking in-the-Wild Challenge: Benchmark and Results

J. Shen, S. Zafeiriou, G. Chrysos, J. Kossaifi, G. Tzimiropoulos, M. Pantic. Proceedings of IEEE International Conference on Computer Vision, 300 Videos in the Wild (300-VW): Facial Landmark Tracking in-the-Wild Challenge & Workshop (ICCVW'15). December 2015.

Bibtex reference [hide]
@inproceedings{300VW,
    author = {J. Shen and S. Zafeiriou and G. Chrysos and J. Kossaifi and G. Tzimiropoulos and M. Pantic},
    booktitle = {Proceedings of IEEE International Conference on Computer Vision, 300 Videos in the Wild (300-VW): Facial Landmark Tracking in-the-Wild Challenge & Workshop (ICCVW'15)},
    month = {December},
    title = {The First Facial Landmark Tracking in-the-Wild Challenge: Benchmark and Results},
    year = {2015},
}
Endnote reference [hide]

Fast and exact Bi-directional Fitting of Active Appearance Models

J. Kossaifi, G. Tzimiropoulos, M. Pantic. Proceedings of the IEEE Int’l Conf. on Image Processing (ICIP’15). Quebec City, QC, Canada, pp. 1135 - 1139, September 2015.

Bibtex reference [hide]
@inproceedings{kossaifi_bidirectional_aam,
    author = {J. Kossaifi and G. Tzimiropoulos and M. Pantic},
    pages = {1135--1139},
    address = {Quebec City, QC, Canada},
    booktitle = {Proceedings of the IEEE Int’l Conf. on Image Processing (ICIP’15)},
    month = {September},
    title = {Fast and exact Bi-directional Fitting of Active Appearance Models},
    year = {2015},
}
Endnote reference [hide]

Fast Newton Active Appearance Models

J. Kossaifi, G. Tzimiropoulos, M. Pantic. Proceedings of the IEEE Int’l Conf. on Image Processing (ICIP’14). Paris, France, pp. 1420 - 1424, October 2014.

Bibtex reference [hide]
@inproceedings{kossaifi_newton_aam,
    author = {J. Kossaifi and G. Tzimiropoulos and M. Pantic},
    pages = {1420--1424},
    address = {Paris, France},
    booktitle = {Proceedings of the IEEE Int’l Conf. on Image Processing (ICIP’14)},
    month = {October},
    title = {Fast Newton Active Appearance Models},
    year = {2014},
}
Endnote reference [hide]