An ILP-based oracle applied to seven VIS methods on YouTube-VIS and OVIS shows tracking instability as the dominant bottleneck, producing gaps exceeding 20 AP under occlusion while classification impact is secondary.
Mask2former for video instance segmentation
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
We find Mask2Former also achieves state-of-the-art performance on video instance segmentation without modifying the architecture, the loss or even the training pipeline. In this report, we show universal image segmentation architectures trivially generalize to video segmentation by directly predicting 3D segmentation volumes. Specifically, Mask2Former sets a new state-of-the-art of 60.4 AP on YouTubeVIS-2019 and 52.6 AP on YouTubeVIS-2021. We believe Mask2Former is also capable of handling video semantic and panoptic segmentation, given its versatility in image segmentation. We hope this will make state-of-the-art video segmentation research more accessible and bring more attention to designing universal image and video segmentation architectures.
citation-role summary
method 1
citation-polarity summary
fields
cs.CV 6roles
method 1polarities
use method 1representative citing papers
A modified video segmentation architecture decouples processing latency from target count, enabling real-time (>36 FPS) tracking of 10+ objects simultaneously while preserving individual identities.
SA-VIS trains video instance segmentation models on sparse frame annotations via a Past-frames Feature Propagation module and frame-specific instance queries, showing only a 0.4% AP drop versus dense training on YouTube-VIS and OVIS benchmarks.
GOLD-BEV learns dense BEV semantic maps including dynamic agents from ego-centric sensors by using synchronized aerial imagery for training supervision and pseudo-label generation.
Primus and PrimusV2 are Transformer-centric models that match or exceed nnU-Net and top CNNs on nine 3D medical segmentation datasets by enforcing attention usage.
PAT-VCM adds lightweight auxiliary tokens to a shared baseline video stream to support multiple downstream machine tasks without task-specific codecs.
citing papers explorer
-
Mind the Gap: Disentangling Performance Bottlenecks in Video Instance Segmentation
An ILP-based oracle applied to seven VIS methods on YouTube-VIS and OVIS shows tracking instability as the dominant bottleneck, producing gaps exceeding 20 AP under occlusion while classification impact is secondary.
-
SAM-MT: Real-Time Interactive Multi-Target Video Segmentation
A modified video segmentation architecture decouples processing latency from target count, enabling real-time (>36 FPS) tracking of 10+ objects simultaneously while preserving individual identities.
-
SA-VIS: Sparse frame Annotations for training Video Instance Segmentation
SA-VIS trains video instance segmentation models on sparse frame annotations via a Past-frames Feature Propagation module and frame-specific instance queries, showing only a 0.4% AP drop versus dense training on YouTube-VIS and OVIS benchmarks.
-
GOLD-BEV: GrOund and aeriaL Data for Dense Semantic BEV Mapping of Dynamic Scenes
GOLD-BEV learns dense BEV semantic maps including dynamic agents from ego-centric sensors by using synchronized aerial imagery for training supervision and pseudo-label generation.
-
Primus: Enforcing Attention Usage for 3D Medical Image Segmentation
Primus and PrimusV2 are Transformer-centric models that match or exceed nnU-Net and top CNNs on nine 3D medical segmentation datasets by enforcing attention usage.
-
PAT-VCM: Plug-and-Play Auxiliary Tokens for Video Coding for Machines
PAT-VCM adds lightweight auxiliary tokens to a shared baseline video stream to support multiple downstream machine tasks without task-specific codecs.