Mind the Gap: Disentangling Performance Bottlenecks in Video Instance Segmentation

• URL: http://arxiv.org/abs/2606.07394v1
• Date: Fri, 05 Jun 2026 15:32:48 GMT
• Title: Mind the Gap: Disentangling Performance Bottlenecks in Video Instance Segmentation
• Abstract summary: Video Instance (VIS) classification, segmentation, and tracking objectives are jointly evaluated.<n>We introduce a diagnostic framework that formulates identity and class assignment as an algorithmic Linear Program (ILP)<n>We also introduce TrackLens, a visual tool that translates magnitude into observable, query-level failure modes.
• Score: 0.34410212782758043
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In Video Instance Segmentation (VIS), classification, segmentation, and tracking objectives are jointly evaluated, but their individual contributions to performance loss remain opaque. We introduce a diagnostic framework that formulates identity and class assignment as an Integer Linear Program (ILP), yielding a model-agnostic oracle that hierarchically isolates each error source. Applied to seven VIS methods spanning online and offline paradigms across YouTube-VIS 2019/2021 and a diagnostic subset of OVIS, our analysis reveals a consistent picture. Tracking instability is a critical bottleneck for online methods, with gaps exceeding 20 AP under heavy occlusion, and grows sharply with video length and instance density. While semantic classification contributes meaningfully on standard benchmarks, its impact becomes negligible where tracking fails most. Although stronger backbones substantially lift default scores, they leave AP tracking gaps largely intact, confirming that temporal fragility is algorithmic rather than purely representational. To complement the oracle, we introduce TrackLens, a visual tool that translates gap magnitude into observable, query-level failure modes. Together, these tools provide a systematic foundation for targeting VIS's core challenge: robust long-term temporal association.

Related papers

• Open-World Video Segmentation [21.65294890698273]
  We introduce Savvy, a practical and strong system for open-world long-horizon video segmentation.<n>Savvy supports persistent object discovery, safe track promotion, and stable long-range identity maintenance.<n>We also propose OGA, a granularity-aware evaluation suite for open-world video segmentation.
  arXiv  Detail & Related papers  (2026-06-14T06:48:15Z)
• Causal Bootstrapped Alignment for Unsupervised Video-Based Visible-Infrared Person Re-Identification [52.784239635604735]
  VVI-ReID is a critical technique for all-day surveillance, where temporal information provides additional cues beyond static images.<n>Existing approaches rely heavily on fully supervised learning with expensive cross-modality annotations, limiting scalability.<n>We propose a Causal Bootstrapped Alignment framework that explicitly exploits inherent video priors.
  arXiv  Detail & Related papers  (2026-04-17T02:15:59Z)
• From Detection to Association: Learning Discriminative Object Embeddings for Multi-Object Tracking [24.903851740154433]
  This paper introduces an explicit feature framework that enhances object discriminativeness across three perspectives.<n>Experiments demonstrate that SpongeBobA achieves state-of-the-art performance on multiple challenging MOT benchmarks, including DanceTrack, SportsMOT, and BFT.
  arXiv  Detail & Related papers  (2025-12-02T04:04:39Z)
• Improving Weakly-supervised Video Instance Segmentation by Leveraging Spatio-temporal Consistency [9.115508086522887]
  We introduce a weakly-supervised method called Eigen VIS that achieves competitive accuracy compared to other VIS approaches. This method is based on two key innovations: a Temporal Eigenvalue Loss (TEL) and a clip-level Quality Co-efficient (QCC) The code is available on https://github.com/farnooshar/EigenVIS.
  arXiv  Detail & Related papers  (2024-08-29T16:05:05Z)
• DVIS-DAQ: Improving Video Segmentation via Dynamic Anchor Queries [60.09774333024783]
  We introduce Dynamic Anchor Queries (DAQ) to shorten the transition gap between the anchor and target queries. We also introduce a query-level object Emergence and Disappearance Simulation (EDS) strategy, which unleashes DAQ's potential without any additional cost. Experiments demonstrate that DVIS-DAQ achieves a new state-of-the-art (SOTA) performance on five mainstream video segmentation benchmarks.
  arXiv  Detail & Related papers  (2024-03-29T17:58:50Z)
• VISAGE: Video Instance Segmentation with Appearance-Guided Enhancement [39.154059294954614]
  Online Video Instance (VIS) methods have shown remarkable advancement with their powerful query-based detectors. However, our observations demonstrate that these methods heavily rely on location information, which often causes incorrect associations between objects. This paper presents that a key axis of object matching in trackers is appearance information, which becomes greatly instructive under conditions where positional cues are insufficient for distinguishing their identities.
  arXiv  Detail & Related papers  (2023-12-08T07:48:03Z)
• Scalable Incomplete Multi-View Clustering with Structure Alignment [71.62781659121092]
  In this paper, we propose a novel incomplete anchor graph learning framework. We construct the view-specific anchor graph to capture the complementary information from different views. The time and space complexity of the proposed SIMVC-SA is proven to be linearly correlated with the number of samples.
  arXiv  Detail & Related papers  (2023-08-31T08:30:26Z)
• CTVIS: Consistent Training for Online Video Instance Segmentation [62.957370691452844]
  Discrimination of instance embeddings plays a vital role in associating instances across time for online video instance segmentation (VIS) Recent online VIS methods leverage CIs sourced from one reference frame only, which we argue is insufficient for learning highly discriminative embeddings. We propose a simple yet effective training strategy, called Consistent Training for Online VIS (CTVIS), which devotes to aligning the training and inference pipelines.
  arXiv  Detail & Related papers  (2023-07-24T08:44:25Z)
• DVIS: Decoupled Video Instance Segmentation Framework [15.571072365208872]
  Video instance segmentation (VIS) is a critical task with diverse applications, including autonomous driving and video editing. Existing methods often underperform on complex and long videos in real world, primarily due to two factors. We propose a decoupling strategy for VIS by dividing it into three independent sub-tasks: segmentation, tracking, and refinement.
  arXiv  Detail & Related papers  (2023-06-06T05:24:15Z)
• Occluded Video Instance Segmentation: Dataset and ICCV 2021 Challenge [133.80567761430584]
  We collect a large-scale dataset called OVIS for video instance segmentation in the occluded scenario. OVIS consists of 296k high-quality instance masks and 901 occluded scenes. All baseline methods encounter a significant performance degradation of about 80% in the heavily occluded object group.
  arXiv  Detail & Related papers  (2021-11-15T17:59:03Z)
• Self-supervised Video Object Segmentation [76.83567326586162]
  The objective of this paper is self-supervised representation learning, with the goal of solving semi-supervised video object segmentation (a.k.a. dense tracking) We make the following contributions: (i) we propose to improve the existing self-supervised approach, with a simple, yet more effective memory mechanism for long-term correspondence matching; (ii) by augmenting the self-supervised approach with an online adaptation module, our method successfully alleviates tracker drifts caused by spatial-temporal discontinuity; (iv) we demonstrate state-of-the-art results among the self-supervised approaches on DAVIS-2017 and YouTube
  arXiv  Detail & Related papers  (2020-06-22T17:55:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.