Saved in:
Bibliographic Details
Main Authors: Dou, Tianyang, Li, Ming, Qin, Jiangying, Liao, Xuan, Zhong, Jiageng, Gruen, Armin, Deng, Mengyi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.16730
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918172169338880
author Dou, Tianyang
Li, Ming
Qin, Jiangying
Liao, Xuan
Zhong, Jiageng
Gruen, Armin
Deng, Mengyi
author_facet Dou, Tianyang
Li, Ming
Qin, Jiangying
Liao, Xuan
Zhong, Jiageng
Gruen, Armin
Deng, Mengyi
contents Coral reefs are vital yet fragile ecosystems that require accurate large-scale mapping for effective conservation. Although global products such as the Allen Coral Atlas provide unprecedented coverage of global coral reef distri-bution, their predictions are frequently limited in spatial precision and semantic consistency, especially in regions requiring fine-grained boundary delineation. To address these challenges, we propose UKANFormer, a novel se-mantic segmentation model designed to achieve high-precision mapping under noisy supervision derived from Allen Coral Atlas. Building upon the UKAN architecture, UKANFormer incorporates a Global-Local Transformer (GL-Trans) block in the decoder, enabling the extraction of both global semantic structures and local boundary details. In experiments, UKANFormer achieved a coral-class IoU of 67.00% and pixel accuracy of 83.98%, outperforming conventional baselines under the same noisy labels setting. Remarkably, the model produces predictions that are visually and structurally more accurate than the noisy labels used for training. These results challenge the notion that data quality directly limits model performance, showing that architectural design can mitigate label noise and sup-port scalable mapping under imperfect supervision. UKANFormer provides a foundation for ecological monitoring where reliable labels are scarce.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16730
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UKANFormer: Noise-Robust Semantic Segmentation for Coral Reef Mapping via a Kolmogorov-Arnold Network-Transformer Hybrid
Dou, Tianyang
Li, Ming
Qin, Jiangying
Liao, Xuan
Zhong, Jiageng
Gruen, Armin
Deng, Mengyi
Computer Vision and Pattern Recognition
Coral reefs are vital yet fragile ecosystems that require accurate large-scale mapping for effective conservation. Although global products such as the Allen Coral Atlas provide unprecedented coverage of global coral reef distri-bution, their predictions are frequently limited in spatial precision and semantic consistency, especially in regions requiring fine-grained boundary delineation. To address these challenges, we propose UKANFormer, a novel se-mantic segmentation model designed to achieve high-precision mapping under noisy supervision derived from Allen Coral Atlas. Building upon the UKAN architecture, UKANFormer incorporates a Global-Local Transformer (GL-Trans) block in the decoder, enabling the extraction of both global semantic structures and local boundary details. In experiments, UKANFormer achieved a coral-class IoU of 67.00% and pixel accuracy of 83.98%, outperforming conventional baselines under the same noisy labels setting. Remarkably, the model produces predictions that are visually and structurally more accurate than the noisy labels used for training. These results challenge the notion that data quality directly limits model performance, showing that architectural design can mitigate label noise and sup-port scalable mapping under imperfect supervision. UKANFormer provides a foundation for ecological monitoring where reliable labels are scarce.
title UKANFormer: Noise-Robust Semantic Segmentation for Coral Reef Mapping via a Kolmogorov-Arnold Network-Transformer Hybrid
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.16730