Singularity-temporal-17M

Reported on 3 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 Vision3 results

Zero-Shot Video RetrievalonActivityNet
text-to-video R@10· uses extra data· 2022-06-07
66.9
best: 92.5 (InternVideo2-6B)
SOTA
Revealing Single Frame Bias for Video-and-Language Learning arXiv:2206.03428
Zero-Shot Video RetrievalonActivityNet
text-to-video R@1· uses extra data· 2022-06-07
30.6
best: 63.2 (InternVideo2-6B)
Revealing Single Frame Bias for Video-and-Language Learning arXiv:2206.03428
Zero-Shot Video RetrievalonActivityNet
text-to-video R@5· uses extra data· 2022-06-07
55.6
best: 85.6 (InternVideo2-6B)
Revealing Single Frame Bias for Video-and-Language Learning arXiv:2206.03428