OST

Reported on 4 benchmarks across 1 task · 1 paper · 1 SOTA

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

Computer Vision4 results

Zero-Shot Action RecognitiononKinetics
Top-5 Accuracy· 2023-11-30
94.6
best: 95.7 (TC-CLIP)
SOTA
OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition arXiv:2312.00096
Zero-Shot Action RecognitiononUCF101
Top-1 Accuracy· 2023-11-30
79.7
best: 92.8 (OTI(ViT-L/14))
OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition arXiv:2312.00096
Zero-Shot Action RecognitiononKinetics
Top-1 Accuracy· 2023-11-30
75.1
best: 78.1 (TC-CLIP)
OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition arXiv:2312.00096
Zero-Shot Action RecognitiononHMDB51
Top-1 Accuracy· 2023-11-30
55.9
best: 64.7 (MOV (ViT-L/14))
OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition arXiv:2312.00096