Text-to-Video Retrieval on MSR-VTT 49 (full-corpus)
24.4R@1InternVideo2-1B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| InternVideo2-1BParams=1412M, Search Pool Size=10,0002026.01 | 24.4 | 43.2 | 52.4 | 2,460 | |
| X-PoolParams=153M, Search Pool Size=10,0002026.01 | 22 | 41 | 51.2 | 3,795 | |
| EERCFParams=164M, Search Pool Size=10,0002026.01 | 20.5 | 42.7 | 50.9 | 4,843 | |
| GRDRParams=63M, Search Pool Size=10,0002026.01 | 17.4 | 32.2 | 39.7 | 184 | |
| T2VIndexerParams=144M, Search Pool Size=10,0002026.01 | 16 | 31.2 | 38.8 | 198 | |
| CLIP4CLIPParams=151M, Search Pool Size=10,0002026.01 | 10.8 | 21.7 | 28.1 | 2,456 | |
| AVGParams=69M, Search Pool Size=10,0002026.01 | 10.3 | 19.8 | 24.3 | 221 | |
| TIGERParams=69M, Search Pool Size=10,0002026.01 | 8.2 | 14.6 | 18.4 | 218 |