InternVideo2.5

Reported on 4 benchmarks across 4 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

VideoonMeViS
J&F· 2025-01-21
32
best: 53.7 (MPG-SAM 2)
InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling arXiv:2501.12386
Instance SegmentationonRefer-YouTube-VOS (2021 public validation)
J&F· 2025-01-21
34.2
best: 73.9 (MPG-SAM 2)
InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling arXiv:2501.12386
Video Object SegmentationonMeViS
J&F· 2025-01-21
32
best: 53.7 (MPG-SAM 2)
InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling arXiv:2501.12386
Referring Expression SegmentationonRefer-YouTube-VOS (2021 public validation)
J&F· 2025-01-21
34.2
best: 73.9 (MPG-SAM 2)
InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling arXiv:2501.12386