Beyond Pixel Overlap: A Framework for Decomposing Segmentation Evaluation Metrics

• URL: http://arxiv.org/abs/2607.00886v1
• Date: Wed, 01 Jul 2026 12:53:25 GMT
• Title: Beyond Pixel Overlap: A Framework for Decomposing Segmentation Evaluation Metrics
• Abstract summary: We treat existing metrics as compositions of modular design choices rather than isolated formulas.<n>We use this framework to analyze representative metrics and show how newer metrics address specific limits in earlier protocols.
• Score: 21.39067846146176
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Evaluation metrics are central to binary target segmentation because they determine how progress is measured, compared, and interpreted. In this paper, target denotes the task-defined positive region to be segmented rather than a generic foreground object. It may be salient, camouflaged, transparent, glass-like, mirror-like, shadow-like, lesion-like, or defined by other application-specific semantics. We treat existing metrics as compositions of modular design choices rather than isolated formulas. The proposed framework decomposes each metric into five stages covering prediction representation, target extraction, target matching, score computation, and metric reporting. We use this framework to analyze representative metrics and show how newer metrics address specific limits in earlier protocols. The stage choices keep each metric's assumptions visible. We then discuss the design space opened by the framework and its implications for task-aware evaluation protocols. Reference code is available at https://github.com/lartpang/PySODMetrics.

Related papers

• Context-measure: Contextualizing Metric for Camouflage [63.31489136704254]
  Camouflage is primarily context-dependent yet current metrics for camouflaged scenarios overlook this critical factor.<n>We propose a new contextualized evaluation paradigm, Context-measure, built upon a probabilistic pixel-aware correlation framework.
  arXiv  Detail & Related papers  (2025-12-08T01:23:28Z)
• Every Component Counts: Rethinking the Measure of Success for Medical Semantic Segmentation in Multi-Instance Segmentation Tasks [60.80828925396154]
  We present Connected-Component(CC)-Metrics, a novel semantic segmentation evaluation protocol. We motivate this setup in the common medical scenario of semantic segmentation in a full-body PET/CT. We show how existing semantic segmentation metrics suffer from a bias towards larger connected components.
  arXiv  Detail & Related papers  (2024-10-24T12:26:05Z)
• LESS: Label-Efficient and Single-Stage Referring 3D Segmentation [55.06002976797879]
  Referring 3D is a visual-language task that segments all points of the specified object from a 3D point cloud described by a sentence of query. We propose a novel Referring 3D pipeline, Label-Efficient and Single-Stage, dubbed LESS, which is only under the supervision of efficient binary mask. We achieve state-of-the-art performance on ScanRefer dataset by surpassing the previous methods about 3.7% mIoU using only binary labels.
  arXiv  Detail & Related papers  (2024-10-17T07:47:41Z)
• A Unified View of Evaluation Metrics for Structured Prediction [41.29492827464339]
  We present a conceptual framework that unifies evaluation metrics for different structured prediction tasks. Our framework requires representing the outputs of these tasks as objects of certain data types. We show that new metrics can be naturally derived in a bottom-up way based on an output structure.
  arXiv  Detail & Related papers  (2023-10-20T20:02:02Z)
• SortedAP: Rethinking evaluation metrics for instance segmentation [8.079566596963632]
  We show that most existing metrics have a limited resolution of segmentation quality. We propose a new metric called sortedAP, which strictly decreases with both object- and pixel-level imperfections.
  arXiv  Detail & Related papers  (2023-09-09T22:50:35Z)
• SILOP: An Automated Framework for Semantic Segmentation Using Image Labels Based on Object Perimeters [11.693197342734152]
  We propose a framework that introduces an additional module using object perimeters for improved saliency. Our new PerimeterFit module will be applied to pre-refine the CAM predictions before using the pixel-similarity-based network. In this way, our PerimeterFit increases the quality of the CAM prediction while simultaneously improving the false negative rate.
  arXiv  Detail & Related papers  (2023-03-14T13:25:55Z)
• RankSEG: A Consistent Ranking-based Framework for Segmentation [5.166970737490847]
  We establish a theoretical foundation of segmentation with respect to the Dice/IoU metrics, including the Bayes rule and Dice-/IoU-calibration. We propose a novel consistent ranking-based framework, namely RankDice/RankIoU, inspired by plug-in rules of the Bayes segmentation rule.
  arXiv  Detail & Related papers  (2022-06-27T07:12:31Z)
• Evaluating Subtitle Segmentation for End-to-end Generation Systems [2.730149977692319]
  Subtitles appear on screen as short pieces of text, segmented based on formal constraints (length) and syntactic/semantic criteria. Standard segmentation metrics cannot be applied when systems generate outputs different than the reference. We introduce $Sigma$, a new Subtitle Score derived from an approximate upper-bound of BLEU on segmentation boundaries.
  arXiv  Detail & Related papers  (2022-05-19T07:38:13Z)
• CoSeg: Cognitively Inspired Unsupervised Generic Event Segmentation [118.18977078626776]
  We propose an end-to-end self-supervised learning framework for event segmentation/boundary detection. Our framework exploits a transformer-based feature reconstruction scheme to detect event boundary by reconstruction errors. The goal of our work is to segment generic events rather than localize some specific ones.
  arXiv  Detail & Related papers  (2021-09-30T14:40:32Z)
• Rethinking Counting and Localization in Crowds:A Purely Point-Based Framework [59.578339075658995]
  We propose a purely point-based framework for joint crowd counting and individual localization. We design an intuitive solution under this framework, which is called Point to Point Network (P2PNet)
  arXiv  Detail & Related papers  (2021-07-27T11:41:50Z)
• Target-Aware Object Discovery and Association for Unsupervised Video Multi-Object Segmentation [79.6596425920849]
  This paper addresses the task of unsupervised video multi-object segmentation. We introduce a novel approach for more accurate and efficient unseen-temporal segmentation. We evaluate the proposed approach on DAVIS$_17$ and YouTube-VIS, and the results demonstrate that it outperforms state-of-the-art methods both in segmentation accuracy and inference speed.
  arXiv  Detail & Related papers  (2021-04-10T14:39:44Z)
• Rethinking Semantic Segmentation Evaluation for Explainability and Model Selection [12.786648212233116]
  We introduce a new metric to assess region-based over- and under-segmentation. We analyze and compare it to other metrics, demonstrating that the use of our metric lends greater explainability to semantic segmentation model performance in real-world applications.
  arXiv  Detail & Related papers  (2021-01-21T03:12:43Z)

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.