Tao Hu's portrait

Tao Hu

Fourth-year PhD, University of Amsterdam, supervised by Cees Snoek and Pascal Mettes

C3.249, Science Park 904, Amsterdam

Email: taohu620 at gmail dot com

 

Publication | GitHub | CV(updated in Jan.2023) |
LinkedIn
| Research Note | Chat with me |Wechat |zhihu

 

Focused on introducing inductive bias into neural network to achieve data-efficiency by few-shot learning, generative model, etc.




News:


[Jan. 2023] Committee members confirmed: Theo Gevers, Marcel Worring, Efstratios Gavves, Bernard Ghanem, Tim Salimans

[Oct. 2022] "Self-Guided Diffusion Models" was accepted by two NeurIPS 2022 workshops.

[Dec. 2021] My mentee Martine Toering's Master Thesis won the ADS Thesis Awards.



Publications

(Full Publication), * Equal contribution

Self-Guided Diffusion Models

Vincent Tao Hu*,David W Zhang*, Yuki M. Asano, Gertjan J. Burghouts, Cees G. M. Snoek

NeurIPS 2022 SSL and SBM workshops.

A bridge between the community of self-supervised learning and diffusion models.

 

Paper | Project | Bibtex

 

Self-supervised Video Representation Learning with Cross-Stream Prototypical Contrasting

Martine Toering, Ioannis Gatopoulos, Maarten Stol, Vincent Tao Hu

WACV 2022

Improve video representation by constrasting Prototypical features.

 

Paper|Code | Bibtex

 

PointMixup: Augmentation for Point Clouds

Yunlu Chen*, Vincent Tao Hu*,Efstratios Gavves, Thomas Mensink, Pascal Mettes, Pengwan Yang, Cees G. M. Snoek

ECCV 2020,Spotlight

A simple augmentation method based on MixUp to boost the performance on related tasks of point cloud.

 

Paper|Code | Bibtex

 

Localizing the Common Action Among a Few Videos

Pengwan Yang*, Vincent Tao Hu*, Pascal Mettes, Cees G. M. Snoek

ECCV 2020

Localizing the temporal extent of an action in a long untrimmed video by attention techniques.

 

Paper| Code | Bibtex

 

SILCO: Show a Few Images, Localize the Common Object

Tao Hu, Pascal Mettes, Jia-Hong Huang, Cees G. M. Snoek

ICCV 2019

Design a graph network and apply attention on them to solve the problem of common object localization.

 

Project | Paper | Code | Poster | Bibtex

 

Attention-based Multi-Context Guiding for Few-Shot Semantic Segmentation

Tao Hu, Pengwan Yang, Chiliang Zhang, Gang Yu, Yadong Mu, Cees G.M. Snoek

AAAI 2019,Spotlight

Solve the few-shot segmentation problem by applying attention in multi-scales.

 

Paper | Supp | Poster | Code | Bibtex

 



Query by Activity Video in the Wild

Tao Hu, William Thong, Pascal Mettes, Cees G.M. Snoek

draft

Few-shot video retrieval.

 

Paper | Bibtex

 

Academic Activities

Reviewer for CVPR 20/21/22/23, ECCV20/22, ICLR22/23, NeurIPS22, AAAI-23 Program Committee Member, BMVC21/22/23, WACV21/22/23, ICRA, TMM, ACCV, etc.

Teaching Assistant for deep learning course(2019/20) with Efstratios Gavves in University of Amsterdam.

Sicily Summer School 2019, Gran Canaria summer school 2021


Supervision

Andrea Rigo(M.S.),2022=>Depth-aware generative modeling

Master Thesis(2022)

Martine Toering(M.S.),2021=>Self-supervised Video Representation Learningwith Cross-Stream Prototypical Contrasting

Master Thesis(2021)

Hinrik Snær(M.S.),2021=>Improvement in efficiency for automated quality control for fish processing using semantic segmentation

Master Thesis(2020)

Martine Toering(B.A.),2019=>Video Classification from scratch

Bachelor Thesis(2019)

Paulien Rouwendaal(M.S.),2019=>Full and point supervised learning strategies for table cell detection

Master Thesis(2019)


Misc

Youtube(For Fun): Vincent Talk

Carl Vondrick, Chen Sun, yuxin wu, Zhifeng Kong




Template is borrowed from Yan-Jun Zhu.

.