修复图像数据集中的镜像填充伪影，提升模型评估可靠性
Finding the Reflection Point: Unpadding Images to Remove Data Augmentation Artifacts in Large Open Source Image Datasets for Machine Learning
- 通过最小均方误差定位镜像边界并去除填充
- 在SHEL5k数据集上使检测平均精度提升至0.61~0.73
- 适合关注数据质量与跨域模型评估的研究者
本文针对机器学习数据集整理中的新问题——检测并消除噪声性镜像填充伪影。尽管填充等数据增强技术用于统一图像尺寸，但可能引入伪影，导致跨领域复用数据集时模型评估失准。我们提出一种系统性算法，基于最小均方误差结合阈值法精确划分反射边界，并移除镜像填充区域。该方法在压缩或插值噪声存在下仍能准确识别真实内容与镜像部分的过渡区。在SHEL5k数据集上的实验表明，使用OWLv2进行零样本目标检测时，安全帽检测的平均精度从0.47升至0.61，人员检测从0.68升至0.73。通过修正标注不一致和填充区扭曲物体，本方法显著提升数据集完整性，实现更可靠的计算机视觉模型评估。
原文摘要 · Abstract (English)
In this paper, we address a novel image restoration problem relevant to machine learning dataset curation: the detection and removal of noisy mirrored padding artifacts. While data augmentation techniques like padding are necessary for standardizing image dimensions, they can introduce artifacts that degrade model evaluation when datasets are repurposed across domains. We propose a systematic algorithm to precisely delineate the reflection boundary through a minimum mean squared error approach with thresholding and remove reflective padding. Our method effectively identifies the transition between authentic content and its mirrored counterpart, even in the presence of compression or interpolation noise. We demonstrate our algorithm's efficacy on the SHEL5k dataset, showing significant performance improvements in zero-shot object detection tasks using OWLv2, with average precision increasing from 0.47 to 0.61 for hard hat detection and from 0.68 to 0.73 for person detection. By addressing annotation inconsistencies and distorted objects in padded regions, our approach enhances dataset integrity, enabling more reliable model evaluation across computer vision tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。