IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Sight …

PhD student at KU Leuven. yztongzhan has 32 repositories available. Follow their code on GitHub.

2022年3月24日 · VideoMAE performs the task of masked video modeling for video pre-training. We propose the extremely high masking ratio (90%-95%) and tube masking strategy to create a challenging task for self-supervised video pre-training. VideoMAE uses the simple masked autoencoder and plain ViT backbone to perform video self-supervised learning.

战昕彤教授多次在国际学术会议及金融业界论坛上获得最佳研究论文奖，并于2021年获颁香港中文大学青年学者研究成就奖 (Young Researcher Award)。 战昕彤教授入选国家海外高层次人才计划青年项目和上海市领军人才项目，并主持国家自然科学基金委面上项目。 她曾连续三年（2018-2021）获得香港研究局的项目资助，包括一项杰出青年研究计划 (ECS) 以及两项优配研究基金 (GRF)，并获得瑞士日内瓦财富研究院，加拿大衍生品研究院等多个海外科研机构的研究支持。

2022年3月23日 · View a PDF of the paper titled VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training, by Zhan Tong and 3 other authors

2023年1月16日 · Zhan Tong, Yibing Song, Jue Wang, Limin Wang Nanjing University, Tencent AI Lab. This repo contains the supported code and scripts to reproduce action detection results of VideoMAE. The code of pre-training is available in original repo.

Zhan Tong's profile on OpenReview, promoting openness in scientific communication and peer-review process.

Zhan Tong Videomae v2: Scaling video masked autoencoders with dual masking Add the full text or supplementary notes for the publication here using Markdown formatting.

Read Zhan Tong's latest research, browse their coauthor's research, and play around with their algorithms