Video Question Answering on HourVideo
35.3AccuracyMMViR
Evaluation Results
| Method | Links | |
|---|---|---|
| MMViRModality=Text+Image, Structure=Multi-grained, Segmentation Type=Clip-based Segmentation, Avg. Time (min)=36.82026.01 | 35.3 | |
| MMViR (w/o img)Modality=Text, Structure=Multi-grained, Segmentation Type=Clip-based Segmentation, Avg. Time (min)=36.02026.01 | 33.6 | |
| MMViRModality=Text+Image, Structure=Multi-grained, Segmentation Type=Kernel Temporal Segmentation, Avg. Time (min)=33.92026.01 | 30.5 | |
| VideoRecap (2024)Modality=Text, Structure=Multi-grained, Avg. Time (min)=67.52026.01 | 29.5 | |
| MMViR (w/o text)Modality=Image, Structure=Fine-grained, Segmentation Type=Clip-based Segmentation, Avg. Time (min)=0.82026.01 | 28.7 | |
| MMViR (w/o text)Modality=Image, Structure=Fine-grained, Segmentation Type=Kernel Temporal Segmentation, Avg. Time (min)=0.82026.01 | 28.2 | |
| Hourvideo (2024)Modality=Text, Structure=Fine-grained, Avg. Time (min)=3.82026.01 | 27.5 | |
| RandomModality=Image, Structure=Global2026.01 | 26.6 | |
| UniformModality=Image, Structure=Global2026.01 | 26.4 | |
| MMViR (w/o img)Modality=Text, Structure=Multi-grained, Segmentation Type=Kernel Temporal Segmentation, Avg. Time (min)=33.12026.01 | 25.2 | |
| MovieChat (2024)Modality=Image, Structure=Global, Avg. Time (min)=158.42026.01 | 19.9 |