PhD student, Huazhong University of Science and Technology, Computer Vision
SeqFormer. SeqFormer: Sequential Transformer for Video Instance Segmentation (ECCV 2022 Oral)
TokBench. Image and video Tokenizer/VAE selection guide, text and face reconstruction evaluation.
cocoapi. COCO API Customized for YouTubeVIS evaluation
wjf5203.github.io. HTML
