SOTAVerified

Video Instance Segmentation

The goal of video instance segmentation is simultaneous detection, segmentation and tracking of instances in videos. In words, it is the first time that the image instance segmentation problem is extended to the video domain.

To facilitate research on this new task, a large-scale benchmark called YouTube-VIS, which consists of 2,883 high-resolution YouTube videos, a 40-category label set and 131k high-quality instance masks is built.

Papers

Showing 76100 of 148 papers

TitleStatusHype
VISOLO: Grid-Based Space-Time Aggregation for Efficient Online Video Instance SegmentationCode1
VITA: Video Instance Segmentation via Object Token AssociationCode1
Object Propagation via Inter-Frame Attentions for Temporally Stable Video Instance SegmentationCode0
One-stage Video Instance Segmentation: From Frame-in Frame-out to Clip-in Clip-outCode0
ThinkVideo: High-Quality Reasoning Video Segmentation with Chain of ThoughtsCode0
PM-VIS+: High-Performance Video Instance Segmentation without Video AnnotationCode0
Robust Online Video Instance Segmentation with Track QueriesCode0
Deformable VisTR: Spatio temporal deformable attention for video instance segmentationCode0
Towards Real-Time Open-Vocabulary Video Instance SegmentationCode0
Efficient Video Object Segmentation via Network ModulationCode0
End-to-end video instance segmentation via spatial-temporal graph neural networksCode0
Video Instance Segmentation in an Open-WorldCode0
False Negative Reduction in Video Instance Segmentation using Uncertainty EstimatesCode0
Foundation Models for Amodal Video Instance Segmentation in Automated DrivingCode0
GRAtt-VIS: Gated Residual Attention for Auto Rectifying Video Instance SegmentationCode0
Two-Level Temporal Relation Model for Online Video Instance SegmentationCode0
Improving Video Instance Segmentation by Light-weight Temporal Uncertainty EstimatesCode0
Learning a Spatio-Temporal Embedding for Video Instance SegmentationCode0
TCOVIS: Temporally Consistent Online Video Instance SegmentationCode0
Temporal RoI Align for Video Object RecognitionCode0
MSN: Efficient Online Mask Selection Network for Video Instance SegmentationCode0
Learning Video Instance Segmentation with Recurrent Graph Neural Networks0
SAM2Auto: Auto Annotation Using FLASH0
SDI-Paste: Synthetic Dynamic Instance Copy-Paste for Video Instance Segmentation0
Deep Learning Techniques for Video Instance Segmentation: A Survey0
Show:102550
← PrevPage 4 of 6Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1CAVIS(VIT-L, Offline)mask AP57.1Unverified
2DVIS-DAQ(VIT-L, Offline)mask AP57.1Unverified
3DVIS++(VIT-L,Offline)mask AP53.4Unverified
4GLEE-Promask AP50.4Unverified
5DVIS(Swin-L, Offline)mask AP49.9Unverified
6DVIS++(VIT-L, Online)mask AP49.6Unverified
7UNINEXT (ViT-H, Online)mask AP49Unverified
8DVIS(Swin-L, Online)mask AP47.1Unverified
9CTVIS (Swin-L)mask AP46.9Unverified
10RefineVIS (Swin-L, offline)mask AP46Unverified
#ModelMetricClaimedVerifiedStatus
1CAVIS(ViT-L, Online)mask AP68.9Unverified
2DVIS++(ViT-L, Online)mask AP67.7Unverified
3DVISmask AP64.9Unverified
4Tube-Linkmask AP64.6Unverified
5MinVIS (Swin-L)mask AP61.6Unverified
6Mask2Former (Swin-L)mask AP60.4Unverified
7UniVS(Swin-L)mask AP60Unverified
8MDQE(Swin-L)mask AP59.9Unverified
9SeqFormer (Swin-L)mask AP59.3Unverified
10DeVIS (Swin-L)mask AP57.1Unverified
#ModelMetricClaimedVerifiedStatus
1CAVIS(VIT-L, Offline)mask AP65.3Unverified
2DVIS-DAQ(VIT-L, Offline)mask AP64.5Unverified
3DVIS++(VIT-L, Offline)mask AP63.9Unverified
4DVIS++(VIT-L, Online)mask AP62.3Unverified
5RefineVIS (Swin-L, online)mask AP61.4Unverified
6GRAtt-VIS (Swin-L)mask AP60.3Unverified
7TarViS (Swin-L)mask AP60.2Unverified
8DVIS(Swin-L)mask AP60.1Unverified
9GenVIS (Swin-L)mask AP60.1Unverified
10NOVIS (Swin-L)mask AP59.8Unverified
#ModelMetricClaimedVerifiedStatus
1DVIS++(VIT-L)mAP_L50.9Unverified
2CAVIS (VIT-L)mAP_L48.6Unverified
3CTVIS (Swin-L)mAP_L46.4Unverified
4DVIS(Swin-L)mAP_L45.9Unverified
5CTVIS (ResNet-50)mAP_L39.4Unverified
6InstanceFormer (Swin)mAP_L26.3Unverified
7InstanceFormer (Resnet-50)mAP_L24.8Unverified
#ModelMetricClaimedVerifiedStatus
1PCANmMOTSA27.4Unverified
2QDTrack-mots-fixmMOTSA23.5Unverified
3QDTrack-motsmMOTSA22.5Unverified
4MaskTrackRCNNmMOTSA12.3Unverified
5STEm-SegmMOTSA12.2Unverified
6SortIoUmMOTSA10.3Unverified
#ModelMetricClaimedVerifiedStatus
1VMT (Swin-L)Tube-Boundary AP44.8Unverified
2SeqFormer (Swin-L)Tube-Boundary AP43.3Unverified
3VMT (R101)Tube-Boundary AP32.5Unverified
4VMT (R50)Tube-Boundary AP30.7Unverified
#ModelMetricClaimedVerifiedStatus
1Temporal ROI Alignmask AP38Unverified
#ModelMetricClaimedVerifiedStatus
1MaskFreeVISAP55.3Unverified