7 papers
DIME: Scaling Facial Representation Learning via Differential Masked Autoencoding
Hao Yu, Haoyu Chen, Hui Wei +3
Scaling up data and model capacity has driven substantial progress in visual representation learning. However, conventional pretext tasks exhibit diminishing returns on fine-graine…
Human-JEPA: A Human-Centric Vision Model that Perceives and Anticipates
Hui Wei, Licai Sun, Guoying Zhao
Machines that understand humans should perceive the present and anticipate the future. Existing human-centric vision model are pretrained on human images, set the state of the art…
Face De-Identification: A Domain-Centric Survey from Capture to Processing
Hui Wei, Hao Yu, Guoying Zhao
Face de-identification (De-ID) aims to remove or conceal personally identifiable facial features in images or videos to prevent identity recognition while preserving utility for do…
CanonicalPhys: Pose-Robust Remote Photoplethysmography via Canonical-Space Priors
Hui Wei, Seyedata Jodeiri Seyedian, Xiaobai Li +1
Deep remote photoplethysmography (rPPG) attains sub-bpm heart-rate error on frontal, stationary faces yet degrades sharply under head pose: on MMPD, the state-of-the-art FactorizeP…
iMiGUE-3K: A Large-Scale Benchmark for Micro-Gesture Analysis with Self-Supervised Learning
Chengyan Wang, Haoyu Chen, Hui Wei +3
Emotion understanding is a fundamental challenge in affective computing and artificial intelligence. While existing approaches predominantly focus on facial expressions and speech,…
Temporally Multi-Scale Sparse Self-Attention for Physical Activity Data Imputation
Hui Wei, Maxwell A. Xu, Colin Samplawski +3
Wearable sensors enable health researchers to continuously collect data pertaining to the physiological state of individuals in real-world settings. However, such data can be subje…
REBAR: Retrieval-Based Reconstruction for Time-series Contrastive Learning
Maxwell A. Xu, Alexander Moreno, Hui Wei +2
The success of self-supervised contrastive learning hinges on identifying positive data pairs, such that when they are pushed together in embedding space, the space encodes useful…