My research interests are in Multi-modality Learning and Data-centric AI. Now I mainly focus on very large-scale efficient vision-language pre-training (VLP). I obtained my bachelor degree and master degree in Sun Yat-Sen University (SYSU). I obtained my PHD degree in National University of Singapore (NUS) and supervised by Prof. Mike Zheng Shou.
News  [Aug 2024] Congratulation!! Our paper "Parrot Captions Teach CLIP to Spot Text" selected as an oral presentation in ECCV24. [June 2024] We release VisInContext, an efficient Interleaved vision-language pre-training framework for increasing the in-context text length.[Jan 2024] We release CosMo, an efficient Interleaved vision-language pre-training framework for both Image and Video! We also release the first interleaved video-text dataset Howto-Interlink7M. See Website for more details. [Dec 2023] We release "Parrot Captions Teach CLIP to Spot Text", an work focus on explore one simple bug in CLIP model. See Website for more details. [Dec 2023] Our extension of "Position-guided Text Prompt for Vision Language Pre-training" is accepted by TPAMI. [April 2023] Our image2paragraph toolbox for fun, Code. [Feb 2023] Our PTP and All-in-one are all accepted by CVPR2023. [Dec 2022] The work "Position-guided Text Prompt for Vision-language Pretraining" has been released in arxiv. Code and pretrained models available here: Code. [Sep 2022] Our first work for large-scale ego-centric pretrained method "Egocentric Video-Language Pretraining" (second-author) has been accepted by NeurIPS 2022. [July 2022] One co-author work has been accepted by ECCV2022. [June 2022] We won first place in CVPR's EPIC-Kitchens Challenge 2022 and OSCC of Ego4D Challenge. Looking kevin's website for details. [Mar 2022] We release the first and simplest e2e one-stream video-language pre-training method: "All in One: Exploring Unified Video-Language Pre-training" in arix! Code and pretrained models are available here: Code. [Mar 2022] Congratulation!! Our paper "Object-aware Video-language Pre-training for Retrieval" has been accepted by CVPR2022! Code. [Mar 2022] Two previous co-author papers has been accepted by ICME. [Feb 2022] One previous co-author paper "Hierarchical Feature Disentangling Network for Universal Domain Adaptation" has been accepted by Pattern Recognition (PR) (IF: 7.2). URL [Nov 2021] Congratulation! Our paper "Suppressing Static Visual Cues via Normalizing Flows for Self-supervised Video Representation Learning" has been accepted by AAAI2022. It will be released in Arxiv soon. [July 2021] Congratulation! Our paper "Learning Spatio-temporal Representation by Channel Aliasing Video Perception" has been accepted by ACM MM. [July 2021] I was awarded the excellent master graduation thesis (1/224). [June 2021] Our paper "Multi-level Temporal Dilated Dense Prediction for Action Recognition" has been published in IEEE Transactions on MultiMedia (TMM) (IF: 6.1). URL [Apr 2021] Congratulation!!! I will take part in Show Lab at the August of this year. [Mar 2021] I was invited to give a talk in VALSE-semanir in 3.28. Weclome to take part in us! Link! [Mar 2021] Our BE was recommended to CVer! Link!> [Mar 2021] I will serve as a reviewer for ICCV2021! [Mar 2021] Our works on video self-supervised learning were accepted to CVPR! [Mar 2021] Our work on self-supervised mutual learning was accepted to ICME! [Jan 2021] I will serve as a reviewer for CVPR2021! [Oct 2020] Our work on using contrastive learning for video action recognition was accepted to AAAI! [Aug 2020] The code for our 3D Net Visualization has been relased in My Github, support no-label visualization. [Apr 2020] Our work on action recognition with dense prediction network was accepted to TCSVT! [Feb 2018] The code for our real time action recognition has been relased in My Github, Publications (First/Co-first Author)  2023.1.1 - Now
  2022.1.1 - 2022.12.31
  2021.1.1 - 2021.12.31
  2020 & before
Awards2022 Showlab Annual Award. ($1,000)The champion of CVPR'22 Epic-kinetics challenge (2022). The 1st place in Ego4D challenge (2022). The final awarded list of AI SINGAPORE PhD FELLOWSHIP PROGRAMME ($$240,000). 2021 Excellent Graduation Thesis (1/224). Reviewer recognitions, CVPR 2021, ICCV 2021, AAAI 2021, TCSVT. The First Prize Scholarship (2020). Second Prize of College Students Innovation and Entrepreneurship Competition (2018). Excellent undergraduate thesis (2017). CollaboratorsI have gotten to work with some wonderful collaborators.@Microsoft Azure AI
Lijuan Wang, Principal Research Manager
Shuicheng Yan, Professor of @NUS
Rongrong Ji, Professor of @XiaMen University
Talk
VisitorsMore informationConfused by the contents of this page? You can contact with my email. |