UP Lab

Universal Perception (UP) Lab is led by Xiatian Zhu, a Senior Lecturer with Surrey Institute for People-Centred Artificial Intelligence, and Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey, Guildford, UK. He was a research scientist at Samsung AI Centre, Cambridge, UK. He received his Ph.D. from the Queen Mary University of London.

The mission of UP Lab is to advance the capabilities of AI through modeling a variety of perception data (e.g. images, videos, audio, text, 3D points). We are working around the understanding of both machine learning theories and real-world domain applications, with the aim to develop transformative technologies for the good of our society.

For research collaboration and studentship openings, please feel free to contact us.

News

Sep 22, 2023	1 NeurIPS paper accepted
Jul 17, 2023	4 ICCV papers and 3 workshop papers accepted
Apr 26, 2023	A survey on multimodal learning with Transformers accepted to IEEE TPAMI
Mar 10, 2023	3 CVPR papers accepted
Jan 7, 2023	Call for papers, First workshop on End-to-End Autonomous Driving: Perception, Prediction, Planning and Simulation, Vancouver, Canada, CVPR 2023

Selected publications

NeurIPS

HeadSculpt: Crafting 3D Head Avatars with Text

Han, Xiao, Cao, Yukang, Han, Kai, Zhu, Xiatian, Deng, Jiankang, Song, Yi-Zhe, Xiang, Tao, and Wong, Kwan-Yee K

In Annual Conference on Neural Information Processing Systems 2023

Bib Code Website

@inproceedings{han2023headsculpt,
  title = {HeadSculpt: Crafting 3D Head Avatars with Text},
  author = {Han, Xiao and Cao, Yukang and Han, Kai and Zhu, Xiatian and Deng, Jiankang and Song, Yi-Zhe and Xiang, Tao and Wong, Kwan-Yee K},
  booktitle = {Annual Conference on Neural Information Processing Systems},
  year = {2023},
}

ICCV

DiffTAD: Temporal Action Detection with Proposal Denoising Diffusion

Nag, Sauradip, Zhu, Xiatian, Deng, Jiankang, Song, Yi-Zhe, and Xiang, Tao

In Proceedings of the IEEE/CVF International Conference on Computer Vision 2023

Bib PDF Code

@inproceedings{nag2023difftad,
  author = {Nag, Sauradip and Zhu, Xiatian and Deng, Jiankang and Song, Yi-Zhe and Xiang, Tao},
  title = {DiffTAD: Temporal Action Detection with Proposal Denoising Diffusion},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  year = {2023},
}

ICCV

Homeomorphism Alignment for Unsupervised Domain Adaptation

Zhou, Lihua, Ye, Mao, Zhu, Xiatian, Xiao, Siying, Fan, Xu-Qian, and Neri, Ferrante

In Proceedings of the IEEE/CVF International Conference on Computer Vision 2023

Bib PDF Supp Code

@inproceedings{zhou2023homeomorphism,
  author = {Zhou, Lihua and Ye, Mao and Zhu, Xiatian and Xiao, Siying and Fan, Xu-Qian and Neri, Ferrante},
  title = {Homeomorphism Alignment for Unsupervised Domain Adaptation},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  year = {2023},
}

ICCV

DeepChange: A Long-Term Person Re-Identification Benchmark with Clothes Change

Xu, Peng, and Zhu, Xiatian

In Proceedings of the IEEE/CVF International Conference on Computer Vision 2023

Bib PDF

@inproceedings{xu2023deepchange,
  author = {Xu, Peng and Zhu, Xiatian},
  booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision},
  title = {DeepChange: A Long-Term Person Re-Identification Benchmark with Clothes Change},
  year = {2023},
}

CVPR

Generative Semantic Segmentation

Chen, Jiaqi, Lu, Jiachen, Zhu, Xiatian, and Zhang, Li

In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2023

Bib PDF Code Poster

@inproceedings{chen2023gss,
  title = {Generative Semantic Segmentation},
  author = {Chen, Jiaqi and Lu, Jiachen and Zhu, Xiatian and Zhang, Li},
  booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year = {2023},
}

CVPR (Highlight)

FAME-ViL: Multi-Tasking Vision-Language Model for Heterogeneous Fashion Tasks

Han, Xiao, Zhu, Xiatian, Yu, Licheng, Zhang, Li, Song, Yi-Zhe, and Xiang, Tao

In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2023

Bib PDF Code Poster

@inproceedings{han2023fame,
  title = {FAME-ViL: Multi-Tasking Vision-Language Model for Heterogeneous Fashion Tasks},
  author = {Han, Xiao and Zhu, Xiatian and Yu, Licheng and Zhang, Li and Song, Yi-Zhe and Xiang, Tao},
  booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year = {2023},
}

CVPR

Post-Processing Temporal Action Detection

Nag, Sauradip, Zhu, Xiatian, Song, Yi-Zhe, and Xiang, Tao

In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2023

Bib PDF Code Poster

@inproceedings{nag2022post,
  author = {Nag, Sauradip and Zhu, Xiatian and Song, Yi-Zhe and Xiang, Tao},
  title = {Post-Processing Temporal Action Detection},
  booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year = {2023},
}

TPAMI

Multimodal Learning with Transformers: A Survey

Xu, Peng, Zhu, Xiatian, and Clifton, David A

IEEE Transactions on Pattern Analysis and Machine Intelligence 2023

Bib PDF

@article{xu2023multimodal,
  author = {Xu, Peng and Zhu, Xiatian and Clifton, David A},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  title = {Multimodal Learning with Transformers: A Survey},
  year = {2023},
}

IJCV

Source-free Domain Adaptation via Target Prediction Distribution Searching

Tang, Song, Chang, An, Fabian, Zhang, Zhu, Xiatian, Ye, Mao, and Zhang, Changshui

International Journal of Computer Vision 2023

Bib PDF Code

@article{tang2023sfda,
  author = {Tang, Song and Chang, An and Fabian, Zhang and Zhu, Xiatian and Ye, Mao and Zhang, Changshui},
  journal = {International Journal of Computer Vision},
  title = {Source-free Domain Adaptation via Target Prediction Distribution Searching},
  year = {2023},
}