提出可扩展的标签错误检测与修正框架，提升目标检测数据集质量。
From Label Error Detection to Correction: A Modular Framework and Benchmark for Object Detection Datasets
- 基于现有检测方法，通过众包微任务复核错误标注
- 在KITTI行人数据上发现18%的漏标或错标
- 仅需少量人力即可修复数百错误，适合数据清洗研究
近年来，目标检测的发展得益于规模日益庞大、类型日益多样的数据集。然而，标签错误会损害数据集质量，影响模型训练与评测结果。尽管已有标签错误检测方法，但大多仅在合成基准或有限人工检查中验证。如何系统性、大规模地纠正错误仍是未解难题。本文提出半自动化纠错框架Rechecked：在现有检测方法的基础上，对错误建议进行轻量级众包微任务复核。我们以KITTI数据集中的行人类别为例，通过众包获取高质量修正标注，发现原始标注中存在18%的缺失或不准确标签。结果表明，结合本框架后，相比从头标注，仅需极少人力即可修复数百错误。但即便最佳方法仍遗漏高达66%的错误，这推动了后续研究，并由我们发布的基准数据集得以实现。
原文摘要 · Abstract (English)
Object detection has advanced rapidly in recent years, driven by increasingly large and diverse datasets. However, label errors often compromise the quality of these datasets and affect the outcomes of training and benchmark evaluations. Although label error detection methods for object detection datasets now exist, they are typically validated only on synthetic benchmarks or via limited manual inspection. How to correct such errors systematically and at scale remains an open problem. We introduce a semi-automated framework for label error correction called Rechecked. Building on existing label error detection methods, their error proposals are reviewed with lightweight, crowd-sourced microtasks. We apply Rechecked to the class pedestrian in the KITTI dataset, for which we crowdsourced high-quality corrected annotations. We detect 18% of missing and inaccurate labels in the original ground truth. We show that current label error detection methods, when combined with our correction framework, can recover hundreds of errors with little human effort compared to annotation from scratch. However, even the best methods still miss up to 66% of the label errors, which motivates further research, now enabled by our released benchmark.
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。