SEMIR: Topology-Preserving Graph Minors for Thin-Structure Segmentation

• URL: http://arxiv.org/abs/2606.24935v1
• Date: Mon, 22 Jun 2026 14:43:31 GMT
• Title: SEMIR: Topology-Preserving Graph Minors for Thin-Structure Segmentation
• Abstract summary: SEMIR is a framework that replaces the pixel lattice with a parameterized graph minor.<n>A lightweight GNN classifies the reduced graph and an exact map lifts predictions to pixel resolution.<n>One pipeline--identical architecture--matches or exceeds domain-specific baselines on TTPLA (power lines), CrackSeg9k (pavement cracks), and SkyScapes Lane (aerial markings) on Dice, IoU, and Boundary F1.
• Score: 1.5222045235700188
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Thin-structure segmentation--power lines, cracks, lane markings at 1-3 pixel width--requires preserving connectivity that standard representations preclude: patching severs continuous structures and conventional superpixels merge thin targets into background before classification. Topology-aware losses penalize connectivity breaks at the objective level but cannot recover what the representation has already destroyed. We propose SEMIR, a framework that replaces the pixel lattice with a parameterized graph minor whose contraction map preserves thin-structure connectivity under the contraction criterion. The minor collapses millions of pixels into tens or hundreds of boundary-aligned supernodes, enabling full-resolution inference without patching at scales demonstrated up to 21 MP in this paper; a lightweight GNN classifies the reduced graph and an exact map lifts predictions to pixel resolution. One pipeline--identical architecture, features, loss, and GNN hyperparameters across all dataset--matches or exceeds domain-specific baselines on TTPLA (power lines), CrackSeg9k (pavement cracks), and SkyScapes Lane (aerial markings) on Dice, IoU, and Boundary F1 while reducing mask fragmentation by at least 4.6x relative to SLIC at matched inference.

Related papers

• Widest-Path Reachability Fields for Connectivity-Preserving Slender Structure Segmentation [3.8331203995580565]
  We propose Widest-Path Reachability Fields (WPRF), a differentiable Max-Min reachability objective that redirects gradient flow to connectivity bottlenecks.<n>WPRF implements a differentiable Max-Min objective via dynamic programming on a domain-restricted graph.<n> Experiments across nine architectures and six datasets validate the bottleneck-focused gradient routing mechanism.
  arXiv  Detail & Related papers  (2026-07-08T08:04:06Z)
• SEMIR: Semantic Minor-Induced Representation Learning on Graphs for Visual Segmentation [1.5222045235700188]
  We introduce SEMIR (Semantic Minor-Induced Representation Learning), a representation framework that decouples inference from the native grid.<n> SEMIR transforms the underlying grid graph into a compact, boundary-aligned graph minor through parameterized edge contraction, node deletion, and edge deletion.<n>We benchmark SEMIR on three tumor segmentation datasets -- BraTS 2021, KiTS23, and LiTS -- where targets exhibit high structural variability and distributional uncertainty.
  arXiv  Detail & Related papers  (2026-05-12T16:52:02Z)
• FMS$^2$: Unified Flow Matching for Segmentation and Synthesis of Thin Structures [1.3111058225843968]
  We propose FMS$2$, a flow-matching framework with two modules.<n>SegFlow is a 2.96M- parameter segmentation model built on a standard encoder-decoder backbone.<n> SynFlow is a mask-conditioned mask $rightarrow$ image generator that produces pixel-aligned synthetic image-mask pairs.
  arXiv  Detail & Related papers  (2026-03-14T00:04:25Z)
• ContextLoss: Context Information for Topology-Preserving Segmentation [2.2742404315918927]
  We propose the novel loss function ContextLoss (CLoss) that improves topological correctness by considering topological errors with their whole context in the critical pixel mask.<n>We benchmark our proposed CLoss on three public datasets (2D & 3D) and our own 3D nano-imaging dataset of bone cement lines.
  arXiv  Detail & Related papers  (2025-06-10T19:09:52Z)
• Hierarchical Superpixel Segmentation via Structural Information Theory [48.488598357738674]
  Superpixel segmentation is a foundation for many higher-level computer vision tasks.<n>We present SIT-HSS, a hierarchical superpixel segmentation method based on structural information theory.<n>We show that SIT-HSS performs better than state-of-the-art unsupervised superpixel segmentation algorithms.
  arXiv  Detail & Related papers  (2025-01-13T05:39:43Z)
• DepGraph: Towards Any Structural Pruning [68.40343338847664]
  We study general structural pruning of arbitrary architecture like CNNs, RNNs, GNNs and Transformers. We propose a general and fully automatic method, emphDependency Graph (DepGraph), to explicitly model the dependency between layers and comprehensively group parameters for pruning. In this work, we extensively evaluate our method on several architectures and tasks, including ResNe(X)t, DenseNet, MobileNet and Vision transformer for images, GAT for graph, DGCNN for 3D point cloud, alongside LSTM for language, and demonstrate that, even with a
  arXiv  Detail & Related papers  (2023-01-30T14:02:33Z)
• PNM: Pixel Null Model for General Image Segmentation [17.971090313814447]
  We present a prior model that weights each pixel according to its probability of being correctly classified by a random segmenter. Experiments on semantic, instance, and panoptic segmentation tasks over three datasets confirm that PNM consistently improves the segmentation quality. We propose a new metric, textitPNM IoU, which perceives the boundary sharpness and better reflects the model segmentation performance in error-prone regions.
  arXiv  Detail & Related papers  (2022-03-13T15:17:41Z)
• PSGR: Pixel-wise Sparse Graph Reasoning for COVID-19 Pneumonia Segmentation in CT Images [83.26057031236965]
  We propose a pixel-wise sparse graph reasoning (PSGR) module to enhance the modeling of long-range dependencies for COVID-19 infected region segmentation in CT images. The PSGR module avoids imprecise pixel-to-node projections and preserves the inherent information of each pixel for global reasoning. The solution has been evaluated against four widely-used segmentation models on three public datasets.
  arXiv  Detail & Related papers  (2021-08-09T04:58:23Z)
• Look Closer to Segment Better: Boundary Patch Refinement for Instance Segmentation [51.59290734837372]
  We propose a conceptually simple yet effective post-processing refinement framework to improve the boundary quality. The proposed BPR framework yields significant improvements over the Mask R-CNN baseline on Cityscapes benchmark. By applying the BPR framework to the PolyTransform + SegFix baseline, we reached 1st place on the Cityscapes leaderboard.
  arXiv  Detail & Related papers  (2021-04-12T07:10:48Z)
• Boundary-Aware Segmentation Network for Mobile and Web Applications [60.815545591314915]
  Boundary-Aware Network (BASNet) is integrated with a predict-refine architecture and a hybrid loss for highly accurate image segmentation. BASNet runs at over 70 fps on a single GPU which benefits many potential real applications. Based on BASNet, we further developed two (close to) commercial applications: AR COPY & PASTE, in which BASNet is augmented reality for "COPY" and "PASTING" real-world objects, and OBJECT CUT, which is a web-based tool for automatic object background removal.
  arXiv  Detail & Related papers  (2021-01-12T19:20:26Z)
• Bending Loss Regularized Network for Nuclei Segmentation in Histopathology Images [69.74667930907314]
  We propose a bending loss regularized network for nuclei segmentation. The proposed bending loss defines high penalties to contour points with large curvatures, and applies small penalties to contour points with small curvature. Minimizing the bending loss can avoid generating contours that encompass multiple nuclei.
  arXiv  Detail & Related papers  (2020-02-03T21:20:50Z)

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.