Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/UniVS(Swin-L)

UniVS(Swin-L)

Reported on 37 benchmarks across 12 tasks · 1 paper · 5 SOTA

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

Computer Vision32 results

Instance SegmentationonDAVIS 2017 (val)
J&F Full video· uses extra data· 2024-02-28
59.4
SOTA
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Referring Expression SegmentationonDAVIS 2017 (val)
J&F Full video· uses extra data· 2024-02-28
59.4
SOTA
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Panoptic SegmentationonVIPSeg
STQ· uses extra data· 2024-02-28
58.2
SOTA
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
VideoonYouTube-VOS 2018
Mean Jaccard & F-Measure· uses extra data· 2024-02-28
71.5
best: 86.9 (XMem (BL30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
VideoonDAVIS 2017 (val)
F-measure· uses extra data· 2024-02-28
79.5
best: 92.6 (XMem (BLK30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
VideoonDAVIS 2017 (val)
Jaccard· uses extra data· 2024-02-28
72.8
best: 86.3 (XMem (BLK30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
VideoonDAVIS 2017 (val)
Mean Jaccard & F-Measure· uses extra data· 2024-02-28
76.2
best: 89.5 (XMem (BLK30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Scene ParsingonVSPW
mIoU· uses extra data· 2024-02-28
59.8
best: 63.8 (DVIS++(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Semantic SegmentationonVSPW
mIoU· uses extra data· 2024-02-28
59.8
best: 63.8 (DVIS++(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Instance SegmentationonRefer-YouTube-VOS (2021 public validation)
F· uses extra data· 2024-02-28
59.5
best: 76.1 (MPG-SAM 2)
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Instance SegmentationonRefer-YouTube-VOS (2021 public validation)
J· uses extra data· 2024-02-28
56.8
best: 71.7 (MPG-SAM 2)
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Instance SegmentationonRefer-YouTube-VOS (2021 public validation)
J&F· uses extra data· 2024-02-28
58
best: 73.9 (MPG-SAM 2)
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Object SegmentationonYouTube-VOS 2018
Mean Jaccard & F-Measure· uses extra data· 2024-02-28
71.5
best: 86.9 (XMem (BL30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Object SegmentationonDAVIS 2017 (val)
F-measure· uses extra data· 2024-02-28
79.5
best: 92.6 (XMem (BLK30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Object SegmentationonDAVIS 2017 (val)
Jaccard· uses extra data· 2024-02-28
72.8
best: 86.3 (XMem (BLK30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Object SegmentationonDAVIS 2017 (val)
Mean Jaccard & F-Measure· uses extra data· 2024-02-28
76.2
best: 89.5 (XMem (BLK30K, MS))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Scene UnderstandingonVSPW
mIoU· uses extra data· 2024-02-28
59.8
best: 63.8 (DVIS++(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Referring Expression SegmentationonRefer-YouTube-VOS (2021 public validation)
F· uses extra data· 2024-02-28
59.5
best: 76.1 (MPG-SAM 2)
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Referring Expression SegmentationonRefer-YouTube-VOS (2021 public validation)
J· uses extra data· 2024-02-28
56.8
best: 71.7 (MPG-SAM 2)
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Referring Expression SegmentationonRefer-YouTube-VOS (2021 public validation)
J&F· uses extra data· 2024-02-28
58
best: 73.9 (MPG-SAM 2)
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS 2021
AP50· uses extra data· 2024-02-28
79.4
best: 87.3 (CAVIS(VIT-L, Offline))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS 2021
AP75· uses extra data· 2024-02-28
63.3
best: 73.2 (CAVIS(VIT-L, Offline))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS 2021
AR1· uses extra data· 2024-02-28
46.2
best: 49.7 (CAVIS(VIT-L, Offline))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS 2021
AR10· uses extra data· 2024-02-28
63.1
best: 70.7 (DVIS-DAQ(VIT-L, Offline))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS 2021
mask AP· uses extra data· 2024-02-28
57.9
best: 65.3 (CAVIS(VIT-L, Offline))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS validation
AP50· uses extra data· 2024-02-28
82.1
best: 89.3 (CAVIS(ViT-L, Online))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS validation
AP75· uses extra data· 2024-02-28
65.3
best: 76.2 (CAVIS(ViT-L, Online))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS validation
AR1· uses extra data· 2024-02-28
54.7
best: 58.3 (CAVIS(ViT-L, Online))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS validation
AR10· uses extra data· 2024-02-28
66.8
best: 73.7 (DVIS++(ViT-L, Online))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonYouTube-VIS validation
mask AP· uses extra data· 2024-02-28
60
best: 68.9 (CAVIS(ViT-L, Online))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Video Instance SegmentationonOVIS validation
mask AP· uses extra data· 2024-02-28
41.7
best: 57.1 (DVIS-DAQ(VIT-L, Offline))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Panoptic SegmentationonVIPSeg
VPQ· uses extra data· 2024-02-28
49.3
best: 58.5 (CAVIS(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115

Audio3 results

10-shot image generationonVIPSeg
STQ· uses extra data· 2024-02-28
58.2
SOTA
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
2D Semantic SegmentationonVSPW
mIoU· uses extra data· 2024-02-28
59.8
best: 63.8 (DVIS++(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
10-shot image generationonVIPSeg
VPQ· uses extra data· 2024-02-28
49.3
best: 58.5 (CAVIS(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115

Medical2 results

Semantic SegmentationonVIPSeg
STQ· uses extra data· 2024-02-28
58.2
SOTA
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115
Semantic SegmentationonVIPSeg
VPQ· uses extra data· 2024-02-28
49.3
best: 58.5 (CAVIS(VIT-L))
UniVS: Unified and Universal Video Segmentation with Prompts as Queries arXiv:2402.18115