Precise Top-Layer Fabric Segmentation for Fabric Destacking with Edge- and Shape-Aware Deep Networks

2026-08-11

AI Summary

This work addresses the challenge of segmenting the topmost layer in stacked fabrics, where ambiguous boundaries and high inter-layer visual similarity hinder accurate delineation. To tackle this issue, the authors propose an edge-aware and shape-aware dual-branch encoder-decoder architecture. The method innovatively integrates an edge detection branch with CAD model–derived shape priors, leveraging a dual-branch supervision mechanism during joint training to simultaneously enhance boundary precision and global shape alignment. Experimental results on a real-world fabric dataset demonstrate that the proposed approach significantly outperforms existing baselines, while ablation studies confirm the effectiveness of the multi-branch design in improving segmentation performance.

Abstract

Fabric destacking requires precise segmentation of the topmost fabric layer, a task complicated by subtle fabric boundaries and high visual similarity between fabric layers. Existing semantic and edge-based segmentation approaches often struggle with these complexities, limiting the performance of robotic manipulation for different tasks. In this work, a novel segmentation training architecture tailored for top-layer fabric segmentation in stacked fabrics is proposed. The method extends the classical encoder-decoder framework by introducing two specialized branches - an edge-aware branch and a shape-aware branch - that are used to supervise the backbone network for better tuning. The edge-aware branch enhances boundary delineation, while the shape-aware branch guides the network to capture and align the overall fabric shape with reference masks derived from Computer Aided Design (CAD) models. Experiments on a real-world fabric dataset demonstrate that the training approach outperforms established baselines, verifying the effectiveness of the multi-branch design through both quantitative results and ablation studies.

Problem

• fabric destacking
• top-layer segmentation
• boundary ambiguity
• visual similarity
• stacked fabrics

Innovation

• edge-aware
• shape-aware
• fabric destacking
• top-layer segmentation
• CAD-guided