Research
My research focuses on egocentric perception and predictive modeling of human behavior. I explore how multi-modal first-person signalsβincluding visual, auditory, and motor cuesβcan be used to understand and anticipate human interactions with objects, people, and environments. The broader goal is to enable intelligent agents to perceive, reason, and act in human-centered, real-world scenarios.
This has led me to explore how egocentric cues can be integrated with 3D human understanding and embodied systems, with potential applications in robotics and AR/VR. If you're interested in chatting or collaborating in this direction, feel free to drop me an email :)
|
News
π½π½π½π½π½π½π½
04.2024   Passed my Qualifier!π§ I will transfer to UIUC in Fall 24. π½π½π½π§π½π½π½
π½π½π½π½π½π½π½
02.2024   One paper accepted by CVPR '24! Check out the AV-CONV!! ππ€ΉββοΈ
01.2023   I'll be interning at Meta Reality Labs in summer 2023! βΎοΈπ©βπ»
08.2022   One journal paper accepted by TNNLS! π
07.2022   One paper accepted by ECCV '22!! πβ‘οΈ
03.2022   Our Ego4D paper has been accepted by CVPR '22 as an oral paper! π₯
|
|
|
SocialGesture: Delving into Multi-person Gesture Understanding
Xu Cao,
Pranav Virupaksha,
Wenqi Jia,
Bolin Lai,
Fiona Ryan,
Sangmin Lee,
James M. Rehg†
CVPR, 2025
paper (coming soon)
|
|
Listen to Look into the Future: Audio-Visual Egocentric Gaze Anticipation
Bolin Lai,
Fiona Ryan,
Wenqi Jia,
Miao Liu†,
James M. Rehg†
ECCV, 2024
project /
paper /
code /
data Split /
supplementary /
video /
poster
|
|
The Audio-Visual Conversational Graph: From an Egocentric-Exocentric Perspective
Wenqi Jia,
Miao Liu,
Hao Jiang,
Ishwarya Ananthabhotla,
James Rehg†,
Vamsi Krishna Ithapu†,
Ruohan Gao†
CVPR, 2024
project /
paper /
code /
dataset (under internal review) /
bibtex
(⇐ Move your cursor on the image for a short demo video!)
|
|
Werewolf Among Us: A Multimodal Dataset for Modeling Persuasion Behaviors in Social Deduction Games
Bolin Lai,
Hongxin Zhang,
Miao Liu,
Aryan Pariani,
Fiona Ryan,
Wenqi Jia,
James Rehg†,
Diyi Yang†
ACL Findings, 2023
project /
paper /
bibtex
|
|
Generative Adversarial Network for Future Hand Segmentation from Egocentric Video
Wenqi Jia,
Miao Liu,
James Rehg†
ECCV, 2022
project /
paper / 
code / 
supplement /
poster /
video / 
bibtex
(⇐ Move your cursor on the image for a short demo video!)
|
|
Ego4D: Around the World in 3,000 Hours of Egocentric Video
Kristen Grauman, et al.
CVPR, 2022   (Oral)
project /
paper /
code /
dataset /
video /
bibtex
|
|
Paying More Attention to Motion:
Attention Distillation for Learning Video Representations
Miao Liu,
Wenqi Jia,
Xin Chen,
Yun Zhang,
Yin Li,
James Rehg†
IJCV, 2021   (Special Issue) (under review)
arXiv /
code /
bibtex
|
|
Holistic-Guided Disentangled Learning With Cross-Video Semantics Mining for Concurrent
First-Person and Third-Person Activity Recognition
Tianshan Liu,
Wenqi Jia,
Rui Zhao,
Kin-Man Lam†,
Jun Kong
TNNLS, 2022
paper /
bibtex
|
Academic Service and Volunteer Work
- Reviewer π, CVPR, ECCV, BMVC, ICCV
- Certified First Responder π, Hong Kong Red Cross, Jun 2015 - Jun 2019
|
|