A Cross-Scale Decoder with Token Refinement for Off-Road Semantic Segmentation

• URL: http://arxiv.org/abs/2603.27931v1
• Date: Mon, 30 Mar 2026 00:59:22 GMT
• Title: A Cross-Scale Decoder with Token Refinement for Off-Road Semantic Segmentation
• Abstract summary: Off-road semantic segmentation is fundamentally challenged by irregular terrain, vegetation clutter, and inherent annotation ambiguity.<n>Existing decoder designs either rely on low-scale bottlenecks that oversmooth fine structural details, or repeatedly fuse high-detail features.<n>We present a cross-scale decoder that explicitly addresses these challenges through three complementary mechanisms.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Off-road semantic segmentation is fundamentally challenged by irregular terrain, vegetation clutter, and inherent annotation ambiguity. Unlike urban scenes with crisp object boundaries, off-road environments exhibit strong class-level similarity among terrain categories, resulting in thick and uncertain transition regions that degrade boundary coherence and destabilize training. Rare or thin structures, such as narrow traversable gaps or isolated obstacles, further receive sparse and unreliable supervision and are easily overwhelmed by dominant background textures. Existing decoder designs either rely on low-scale bottlenecks that oversmooth fine structural details, or repeatedly fuse high-detail features, which tends to amplify annotation noise and incur substantial computational cost. We present a cross-scale decoder that explicitly addresses these challenges through three complementary mechanisms. First, a global--local token refinement module consolidates semantic context on a compact bottleneck lattice, guided by boundary-aware regularization to remain robust under ambiguous supervision. Second, a gated detail bridge selectively injects fine-scale structural cues only once through cross-scale attention, preserving boundary and texture information while avoiding noise accumulation. Third, an uncertainty-guided class-aware point refinement selectively updates the least reliable pixels, improving rare and ambiguous structures with minimal computational overhead. The resulting framework achieves noise-robust and boundary-preserving segmentation tailored to off-road environments, recovering fine structural details while maintaining deployment-friendly efficiency. Experimental results on standard off-road benchmarks demonstrate consistent improvements over prior approaches without resorting to heavy dense feature fusion.

Related papers

• Breaking Structural Isolation: Scalable Graph Clustering via Community-Aware Sampling and Structural Entropy [57.058369251551795]
  Unsupervised graph clustering is a fundamental technique for uncovering underlying semantic patterns in large-scale networks.<n>We propose SCISE, a framework that preserves structural Integrity by synergizing community-aware sampling with constrained Structural Entropy.<n>We show that SCISE significantly outperforms state-of-the-art algorithms, with ablation studies and robustness analyses further validating its effectiveness and reliability for real-world large-scale graphs.
  arXiv  Detail & Related papers  (2026-07-06T07:55:53Z)
• Implicit Fuzzification via Bounded Noise Injection for Robust Medical Image Segmentation [2.9974998694295087]
  NoiseUNet injects bounded perturbations into skip connections to regularize cross-scale feature fusion.<n>Experiments demonstrate that NoiseUNet consistently improves both segmentation accuracy and boundary fidelity.
  arXiv  Detail & Related papers  (2026-06-03T04:17:15Z)
• AnchorMoE: Interpretable Time Series Classification via Anchor-Routed MoE [51.994752158953084]
  AnchorMoE is an interpretable-by-construction classification framework.<n>It encodes multi-view representations of local patches and routes them to specialized experts.<n>Experiments on real-world and synthetic benchmarks demonstrate that AnchorMoE achieves highly competitive classification performance.
  arXiv  Detail & Related papers  (2026-06-02T13:30:54Z)
• Hierarchical Consistency Learning for Test-time Adaptation in Camouflage Perception [50.278200968044665]
  Camouflaged object detection (COD) aims to localize targets that exhibit minimal perceptual differences from backgrounds through physical attributes.<n>Existing methods, constrained by the static train-then-freeze paradigm, suffer from domain rigidity and dependency annotation.<n>We propose the hierarchical consistency learning framework, which integrates test-time adaptation for dynamic representation recalibration.
  arXiv  Detail & Related papers  (2026-05-25T09:57:46Z)
• OcclusionFormer: Arranging Z-Order for Layout-Grounded Image Generation [49.49933361280724]
  We introduce OcclusionFormer, a novel framework that explicitly models Z-order priority by decoupling instances.<n>We also introduce a queried alignment loss that explicitly supervises individual instances and enhances semantic consistency.<n>The proposed method effectively reduces ambiguity in overlapping regions, enforces correct occlusion dependencies, and preserves structural integrity, leading to substantial accuracy gains across diverse scenes.
  arXiv  Detail & Related papers  (2026-05-20T16:10:44Z)
• SpectraFlow: Unifying Structural Pretraining and Frequency Adaptation for Medical Image Segmentation [3.182068203318624]
  Recent self-supervised pretraining has improved transferability, but it often exhibits a texture bias.<n>We propose a two-stage framework that couples structure-aware encoder pretraining with boundary-oriented decoding.<n> Experiments on ISIC-2016, Kvasir-SEG, and GlaS demonstrate consistent gains over state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-05-14T08:37:10Z)
• Weakly supervised multimodal segmentation of acoustic borehole images with depth-aware cross-attention [0.0]
  We introduce a weakly supervised multimodal segmentation framework that refines threshold-guided pseudo-labels through learned models.<n>We establish that threshold-guided learned refinement provides the most robust improvement over raw thresholding.<n>The strongest model, confidence-gated depth-aware cross-attention (CG-DCA), consistently outperforms threshold-based, image-only, and earlier multimodal baselines.
  arXiv  Detail & Related papers  (2026-03-21T09:28:05Z)
• Rethinking Vector Field Learning for Generative Segmentation [50.08025820235397]
  Taming diffusion models for generative segmentation has attracted increasing attention.<n>We revisit diffusion segmentation from the perspective of vector field learning.<n>We propose a principled vector field reshaping strategy that augments the learned vanishing velocity field with a detached distance-aware correction term.<n>This correction introduces both attractive and repulsive interactions, enhancing gradient magnitudes near centroids while preserving the original diffusion training framework.
  arXiv  Detail & Related papers  (2026-03-19T17:58:19Z)
• CroBIM-U: Uncertainty-Driven Referring Remote Sensing Image Segmentation [8.834663340762562]
  Referring remote sensing image segmentation aims to localize specific targets described by natural language within complex overhead imagery.<n>Existing methods typically employ uniform fusion and refinement strategies across the entire image.<n>We propose an textbfuncertainty-guided framework that explicitly leverages a pixel-wise Referrbfreferring uncertainty map as a spatial prior to orchestrate adaptive inference.
  arXiv  Detail & Related papers  (2026-01-07T01:02:39Z)
• Lightweight Transformer Framework for Weakly Supervised Semantic Segmentation [0.45880283710344055]
  Weakly supervised semantic segmentation (WSSS) must learn dense masks from noisy, under-specified cues.<n>We show that three small, synergistic changes make weak supervision markedly more effective without altering the MiT backbone.<n>Our method, CrispFormer, augments the decoder with: (1) a boundary branch that supervises thin object contours using a lightweight edge head and a boundary-aware loss; (2) an uncertainty-guided refiner that predicts per-pixel aleatoric uncertainty and uses it to weight losses and gate a residual correction of the segmentation logits; and (3) a dynamic multi-scale fusion layer that replaces static concate
  arXiv  Detail & Related papers  (2025-11-24T22:40:57Z)
• MetaDCSeg: Robust Medical Image Segmentation via Meta Dynamic Center Weighting [77.31583168890633]
  Medical image segmentation is crucial for clinical applications, but it is frequently disrupted by noisy annotations and ambiguous anatomical boundaries.<n>We propose MetaDCSeg, a robust framework that learns optimal pixel-wise weights to suppress the influence of noisy ground-truth labels.<n>Our approach utilizes weighted feature distances for foreground, background, and boundary centers, directing the model's attention toward hard-to-segment pixels near ambiguous boundaries.
  arXiv  Detail & Related papers  (2025-11-24T08:51:02Z)
• Terrain-Enhanced Resolution-aware Refinement Attention for Off-Road Segmentation [0.7734726150561086]
  Designs that fuse only at low resolution blur edges and propagate local errors.<n>We introduce a resolutionaware token decoder that balances global semantics, local consistency, and boundary fidelity under imperfect supervision.
  arXiv  Detail & Related papers  (2025-11-03T10:36:57Z)
• Temporally Consistent Unsupervised Segmentation for Mobile Robot Perception [1.1749564892273827]
  We introduce Frontier-Seg, a method for temporally consistent unsupervised segmentation of terrain from mobile robot video streams.<n>We evaluate Frontier-Seg on a diverse set of benchmark datasets-including RUGD and RELLIS-3D-demonstrating its ability to perform unsupervised segmentation across unstructured off-road environments.
  arXiv  Detail & Related papers  (2025-07-29T19:41:37Z)
• PFSD: A Multi-Modal Pedestrian-Focus Scene Dataset for Rich Tasks in Semi-Structured Environments [73.80718037070773]
  We present the multi-modal Pedestrian-Focused Scene dataset, rigorously annotated in semi-structured scenes with the format of nuScenes.<n>We also propose a novel Hybrid Multi-Scale Fusion Network (HMFN) to detect pedestrians in densely populated and occluded scenarios.
  arXiv  Detail & Related papers  (2025-02-21T09:57:53Z)
• Mesh Denoising Transformer [104.5404564075393]
  Mesh denoising is aimed at removing noise from input meshes while preserving their feature structures. SurfaceFormer is a pioneering Transformer-based mesh denoising framework. New representation known as Local Surface Descriptor captures local geometric intricacies. Denoising Transformer module receives the multimodal information and achieves efficient global feature aggregation.
  arXiv  Detail & Related papers  (2024-05-10T15:27:43Z)
• Progressive Feature Self-reinforcement for Weakly Supervised Semantic Segmentation [55.69128107473125]
  We propose a single-stage approach for Weakly Supervised Semantic (WSSS) with image-level labels. We adaptively partition the image content into deterministic regions (e.g., confident foreground and background) and uncertain regions (e.g., object boundaries and misclassified categories) for separate processing. Building upon this, we introduce a complementary self-enhancement method that constrains the semantic consistency between these confident regions and an augmented image with the same class labels.
  arXiv  Detail & Related papers  (2023-12-14T13:21:52Z)

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.