List your company

Text-to-Video Retrieval on MSVD (val)

51.8Recall@1Cap4Video
18.5227.1635.844.44Dec 31, 2022Feb 15, 2023Apr 3, 2023May 20, 2023Jul 5, 2023Aug 21, 2023Oct 7, 2023

Evaluation results

MethodLinks
Cap4Video2022.1251.880.888.318.3
X-Pool2022.1247.277.48629.3
CLIP4Clip2022.1245.275.584.3210.3
HowToCaptionVideo-Text Training Data=HowToCaption (ours), Evaluation Protocol=Zero-shot2023.1044.573.382.12—
WebVid2MVideo-Text Training Data=WebVid2M [4], Evaluation Protocol=Zero-shot2023.1044.573.482.12—
VideoCC3MVideo-Text Training Data=VideoCC3M [39], Evaluation Protocol=Zero-shot2023.1039.666.776.82—
HowTo100M (ASRs)Video-Text Training Data=HowTo100M with ASRs [17], Evaluation Protocol=Zero-shot2023.1039.268.377.12—
HTM-AAVideo-Text Training Data=HTM-AA (auto-aligned ASRs) [17], Evaluation Protocol=Zero-shot2023.1038.768.678.72—
BLIP InitializationVideo-Text Training Data=None (BLIP initialization), Evaluation Protocol=Zero-shot2023.1038.565742—
HowTo100M (distant supervision)Video-Text Training Data=HowTo100M with distant supervision [31], Evaluation Protocol=Zero-shot2023.1038.568.679.42—
CLIP2022.123764.173.83—
TMVM2022.1236.767.481.32.5—
Frozen2022.1233.764.776.33—
SUPPORT2022.1228.46072.94—
CE2022.1219.84963.86—