Guardado en:
Detalles Bibliográficos
Autores principales: Yu, Yingda, Xuan, Jiaqi, Shi, Shuhui, Teng, Xuanyu, Xu, Shuyang, Tong, Guanchao
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2601.11640
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909993122398208
author Yu, Yingda
Xuan, Jiaqi
Shi, Shuhui
Teng, Xuanyu
Xu, Shuyang
Tong, Guanchao
author_facet Yu, Yingda
Xuan, Jiaqi
Shi, Shuhui
Teng, Xuanyu
Xu, Shuyang
Tong, Guanchao
contents Agricultural weed detection on edge devices is subject to strict constraints on model capacity, computational resources, and real-time inference latency, which prevent performance improvements through model scaling or ensembling. This paper proposes Model-Driven Data Correction (MDDC), a data-centric framework that enhances detection performance by iteratively diagnosing and correcting data quality deficiencies. An automated error analysis procedure categorizes detection failures into four types: false negatives, false positives, class confusion, and localization errors. These error patterns are systematically addressed through a structured train-fix-retrain pipeline with version-controlled data management. Experimental results on multiple weed detection datasets demonstrate consistent improvements of 5-25 percent in mAP at 0.5 using a fixed lightweight detector (YOLOv8n), indicating that systematic data quality optimization can effectively alleviate performance bottlenecks under fixed model capacity constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11640
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Confident Learning for Object Detection under Model Constraints
Yu, Yingda
Xuan, Jiaqi
Shi, Shuhui
Teng, Xuanyu
Xu, Shuyang
Tong, Guanchao
Computer Vision and Pattern Recognition
Agricultural weed detection on edge devices is subject to strict constraints on model capacity, computational resources, and real-time inference latency, which prevent performance improvements through model scaling or ensembling. This paper proposes Model-Driven Data Correction (MDDC), a data-centric framework that enhances detection performance by iteratively diagnosing and correcting data quality deficiencies. An automated error analysis procedure categorizes detection failures into four types: false negatives, false positives, class confusion, and localization errors. These error patterns are systematically addressed through a structured train-fix-retrain pipeline with version-controlled data management. Experimental results on multiple weed detection datasets demonstrate consistent improvements of 5-25 percent in mAP at 0.5 using a fixed lightweight detector (YOLOv8n), indicating that systematic data quality optimization can effectively alleviate performance bottlenecks under fixed model capacity constraints.
title Confident Learning for Object Detection under Model Constraints
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.11640