AI Paper Atlas

Research map / Benchmarks / YouTube-VIS

YouTube-VIS leaderboard

14 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on YouTube-VIS. Best AP (val): 47.6 (Propose-Reduce (Seq Mask R-CNN), ICCV 2021).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

YouTube-VIS · AP (val)

14 papers, each with its best reported number for its own method; higher is better.

#AP ↑MethodPaperTopic
147.6Propose-Reduce (Seq Mask R-CNN)
ResNeXt-101
Video Instance Segmentation With a Propose-Reduce Paradigm
ICCV 2021 · Table 1
Video segmentation and tracking
240.1VisTR
ResNet-101
End-to-End Video Instance Segmentation With Transformers
CVPR 2021 · Main Results table
Video segmentation and tracking
338.3Ours*
SOLO
Learning to Track Instances without Video Annotations
CVPR 2021 · Table: Comparison with SOTA me
Video segmentation and tracking
438MT R-CNN + TROI
X101-FPN
Temporal ROI Align for Video Object Recognition
AAAI 2021 · Table: Applying Temporal ROI A
Video segmentation and tracking
537.6PCAN
ResNet-101
Prototypical Cross-Attention Networks for Multiple Object Tracking and Segmentation
NeurIPS 2021 · Table 1
Tracking
636.8STMask (ada)
ResNet-101-DCN
Spatial Feature Calibration and Temporal Fusion for Effective One-Stage Video Instance Segmentation
CVPR 2021 · Table: Quantitative performanc
Video segmentation and tracking
736.5VisSTG
ResNet-50
End-to-End Video Instance Segmentation via Spatial-Temporal Graph Neural Networks
ICCV 2021 · Table 3
Video segmentation and tracking
836.3SG-Net
ResNet-101
SG-Net: Spatial Granularity Network for One-Stage Video Instance Segmentation
CVPR 2021 · Table 1
CNN features
936.2QueryInst-VIS* (multi-scale aug.)
ResNet-50
Instances As Queries
ICCV 2021 · Table 3
CNN features
1035.3CompFeat
MaskTrack R-CNN
CompFeat: Comprehensive Feature Aggregation for Video Instance Segmentation
AAAI 2021 · Comparison with the State-of-t
Video segmentation and tracking
4 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.