VicTR (ViT-B/16)

Reported on 2 benchmarks across 1 task · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision2 results

Zero-Shot Action RecognitiononUCF101
Top-1 Accuracy· 2023-04-05
72.4
best: 92.8 (OTI(ViT-L/14))
VicTR: Video-conditioned Text Representations for Activity Recognition arXiv:2304.02560
Zero-Shot Action RecognitiononHMDB51
Top-1 Accuracy· 2023-04-05
51
best: 64.7 (MOV (ViT-L/14))
VicTR: Video-conditioned Text Representations for Activity Recognition arXiv:2304.02560