Home

Awesome

Rethinking CLIP-based Video Learners in Cross-Domain Open-Vocabulary Action Recognition

🌈 Overview

alt text

<!-- https://github.com/KunyuLin/XOV-Action/blob/main/xovaction_setting.png?raw=true -->

alt text

<!-- https://github.com/KunyuLin/XOV-Action/blob/main/xovaction_results.png?raw=true -->

📚 XOV-Action Benchmark

New Features 🔥

Benchmark Components

Training Datasets

Test Datasets

Evaluation Metrics

Please refer to our paper for more details.

🚀 Methodology

📌 Acknowledgement

@misc{lin2024xovaction,
  author       = {Kun-Yu Lin, Henghui Ding, Jiaming Zhou, Yi-Xing Peng, Zhilin Zhao, Chen Change Loy, Wei-Shi Zheng},
  title        = {Rethinking CLIP-based Video Learners in Cross-Domain Open-Vocabulary Action Recognition},
  year         = {2024},
  eprint       = {2403.01560},
  archivePrefix= {arXiv},
  primaryClass = {cs.CV}
}