| action-recognition-in-videos-on-hmdb-51 | InternVideo | #1 | Average accuracy of 3 splits: 89.3 |
| action-recognition-in-videos-on-something-something-v1 | InternVideo | #1 | Top 1 Accuracy: 70.0 |
| action-recognition-in-videos-on-something-something-v2 | InternVideo | #2 | Top-1 Accuracy: 77.2 |
| spatio-temporal-action-localization-on-ava-kinetics | InternVideo | #2 | val mAP: 41.01 |
| video-retrieval-on-msvd | InternVideo | #2 | text-to-video R@1: 58.4video-to-text R@1: 76.3 |
| action-classification-on-kinetics-400 | InternVideo | #3 | Acc@1: 91.1 |
| action-classification-on-kinetics-700 | InternVideo-T | #3 | Top-1 Accuracy: 84.0 |
| video-question-answering-on-star | InternVideo | #4 | Average Accuracy: 58.7 |
| zero-shot-video-question-answer-on-star | InternVideo | #4 | Accuracy: 41.6 |
| action-classification-on-kinetics-600 | InternVideo-T | #5 | Top-1 Accuracy: 91.3 |
| action-recognition-on-ava-v2-2 | InternVideo | #5 | mAP: 41.01 |
| temporal-action-localization-on-thumos14 | ActionFormer (InternVideo features) | #5 | Avg mAP (0.3:0.7): 71.58 |
| visual-question-answering-on-msrvtt-qa | InternVideo | #5 | Accuracy: 0.471 |
| zero-shot-action-recognition-on-kinetics-400-zero-shot-action-classification | InternVideo (zero-shot) | #5 | Accuracy: 64.25 |
| zero-shot-video-retrieval-on-vatex | InternVideo | #5 | text-to-video R@1: 49.5video-to-text R@1: 69.5 |
| temporal-action-localization-on-fineaction | InternVideo | #6 | mAP: 17.57 |
| video-retrieval-on-vatex | InternVideo | #6 | text-to-video R@1: 71.1video-to-text R@1: 87.2 |
| temporal-action-localization-on-hacs | InternVideo | #7 | Average-mAP: 41.55 |
| video-retrieval-on-lsmdc | InternVideo | #7 | text-to-video R@1: 34.0video-to-text R@1: 34.9 |
| zero-shot-video-retrieval-on-lsmdc | InternVideo | #7 | text-to-video R@1: 17.6text-to-video R@5: 32.4… |
| video-retrieval-on-activitynet | InternVideo | #8 | text-to-video R@1: 62.2video-to-text R@1: 62.8 |
| zero-shot-video-question-answer-on-tvqa | InternVideo (no speech) | #8 | Accuracy: 35.9 |
| temporal-action-localization-on-activitynet-1-3 | InternVideo | #9 | mAP: 39.00 |
| video-retrieval-on-msr-vtt | InternVideo | #9 | text-to-video R@1: 55.2video-to-text R@1: 57.9 |
| visual-question-answering-on-msvd-qa | InternVideo | #9 | Accuracy: 0.555 |
| video-retrieval-on-didemo | InternVideo | #10 | text-to-video R@1: 57.9video-to-text R@1: 59.1 |
| zero-shot-video-retrieval-on-activitynet | InternVideo | #10 | text-to-video R@1: 30.7video-to-text R@1: 31.4 |
| zero-shot-video-retrieval-on-msvd | InternVideo | #11 | text-to-video R@1: 43.4video-to-text R@1: 67.6 |
| zero-shot-video-retrieval-on-didemo | InternVideo | #16 | text-to-video R@1: 31.5text-to-video R@5: 57.6… |
| zero-shot-video-retrieval-on-msr-vtt | InternVideo | #17 | text-to-video R@1: 40.7video-to-text R@1: 39.6 |
| zero-shot-video-question-answer-on-egoschema-fullset | InternVideo | #51 | Accuracy: 32.1 |