DBI Backend for Nintendo Switch (中文支持)

Hyperbolic Visual Embedding Learning for Zero-Shot Recognition (CVPR 2020)

[WACV'25 Oral] Enhancing Zero-Shot Facial Expression Recognition by LLM Knowledge Transfer

Code for our Source-free Unsupervised Video Domain Adaptation Paper

[AAAI'25]: Building a Multi-modal Spatiotemporal Expert for Zero-shot Action Recognition with CLIP

Official implementation of the paper Transferable-guided Attention Is All You Need for Video Domain Adaptation

A unified source-free domain adaptation framework.

[AAAI 2024 Oral] M2CLIP: A Multimodal, Multi-Task Adapting Framework for Video Action Recognition

[CVPR 2023] Official repository of paper titled "Fine-tuned CLIP models are efficient video learners".

This is the official implement of paper "ActionCLIP: A New Paradigm for Action Recognition"

Out-of-Distribution Detection for Generalized Zero-Shot Action Recognition

【CVPR'2023】Bidirectional Cross-Modal Knowledge Exploration for Video Recognition with Pre-trained Vision-Language Models

Official code for Tell Me What You See: A Zero-Shot Action Recognition Method Based on Natural Language Descriptions (Multimedia Tools and Applications 2024)

[CVPR'25] EMOE: Modality-Specific Enhanced Dynamic Emotion Experts

Accepted by ACM MM 2024, also ACM MM 2024 Grand Challenge Micro-Action Analysis Track-1 Top 1 solution.

Official implementation of "Difficulty-aware Balancing Margin Loss for Long-Tailed Recognition" (AAAI 2025)

[AAAI 2025] Official implementation of the paper: Prototypical Calibrating Ambiguous Samples for Micro-Action Recognition

[NeurIPS 2023] Unsupervised Video Domain Adaptation for Action Recognition: A Disentanglement Perspective

SF(DA)²: Source-free Domain Adaptation Through the Lens of Data Augmentation (ICLR 2024)

[ECCV 2024] Official Implementation of CoPT: Unsupervised Domain Adaptive Segmentation using Domain-Agnostic Text Embeddings

The Pytorch implementation of Domain-Agnostic Mutual Prompting for Unsupervised Domain Adaptation

[CVPR 2024] Code for UNITE, an unsupervised approach for video domain adaptation (https://arxiv.org/abs/2312.02914)

[NeurIPS 2022 Spotlight] VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

[MAC 2024] The baseline code for MAC 2024.