Saved in:
Bibliographic Details
Main Author: Panboonyuen, Teerapong
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.10528
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912426773970944
author Panboonyuen, Teerapong
author_facet Panboonyuen, Teerapong
contents We present SLICK, a novel framework for precise and robust car damage segmentation that leverages structural priors and domain knowledge to tackle real-world automotive inspection challenges. SLICK introduces five key components: (1) Selective Part Segmentation using a high-resolution semantic backbone guided by structural priors to achieve surgical accuracy in segmenting vehicle parts even under occlusion, deformation, or paint loss; (2) Localization-Aware Attention blocks that dynamically focus on damaged regions, enhancing fine-grained damage detection in cluttered and complex street scenes; (3) an Instance-Sensitive Refinement head that leverages panoptic cues and shape priors to disentangle overlapping or adjacent parts, enabling precise boundary alignment; (4) Cross-Channel Calibration through multi-scale channel attention that amplifies subtle damage signals such as scratches and dents while suppressing noise like reflections and decals; and (5) a Knowledge Fusion Module that integrates synthetic crash data, part geometry, and real-world insurance datasets to improve generalization and handle rare cases effectively. Experiments on large-scale automotive datasets demonstrate SLICK's superior segmentation performance, robustness, and practical applicability for insurance and automotive inspection workflows.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10528
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SLICK: Selective Localization and Instance Calibration for Knowledge-Enhanced Car Damage Segmentation in Automotive Insurance
Panboonyuen, Teerapong
Computer Vision and Pattern Recognition
We present SLICK, a novel framework for precise and robust car damage segmentation that leverages structural priors and domain knowledge to tackle real-world automotive inspection challenges. SLICK introduces five key components: (1) Selective Part Segmentation using a high-resolution semantic backbone guided by structural priors to achieve surgical accuracy in segmenting vehicle parts even under occlusion, deformation, or paint loss; (2) Localization-Aware Attention blocks that dynamically focus on damaged regions, enhancing fine-grained damage detection in cluttered and complex street scenes; (3) an Instance-Sensitive Refinement head that leverages panoptic cues and shape priors to disentangle overlapping or adjacent parts, enabling precise boundary alignment; (4) Cross-Channel Calibration through multi-scale channel attention that amplifies subtle damage signals such as scratches and dents while suppressing noise like reflections and decals; and (5) a Knowledge Fusion Module that integrates synthetic crash data, part geometry, and real-world insurance datasets to improve generalization and handle rare cases effectively. Experiments on large-scale automotive datasets demonstrate SLICK's superior segmentation performance, robustness, and practical applicability for insurance and automotive inspection workflows.
title SLICK: Selective Localization and Instance Calibration for Knowledge-Enhanced Car Damage Segmentation in Automotive Insurance
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.10528