视觉目标检测中的信息密度失衡
Information Density Imbalance in Visual Object Detection
另 3 家 · 查看机构详情
- Technical University of Munich（慕尼黑工业大学）
- Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学高瓴人工智能学院）
- Nanyang Technological University（南洋理工大学）
- Baidu Inc.（百度公司）
- Beijing Academy of Artificial Intelligence（北京人工智能研究院）
- University of Manchester（曼彻斯特大学）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
本研究针对目标检测中仅用实例数量无法解释的类别偏差问题，引入信息密度概念，改进三种损失函数，在多数据集上验证其可降低偏差并提升性能，为相关研究提供新视角与工具。
中文摘要 AI 辅助
在目标检测中，通常用实例数量判断数据集是否呈现长尾分布，隐含假设模型在实例较少的类别上表现较差，这一假设催生了大量关于实例数量失衡数据集类别偏差的研究。然而，即便在实例数量相对均衡的数据集上，模型仍会表现出类别偏差，说明仅用实例数量无法解释该现象。本研究首先引入信息密度的概念及其测量方法，随后观察到类别信息密度与准确率存在显著负相关，并探究训练过程对该关系的影响。实证研究表明，信息密度失衡可能是类别偏差的潜在来源。为初步验证信息密度的潜力，我们用该概念对三种先进的目标检测损失函数进行了简单改进。在Pascal VOC、COCO-LT和LVIS数据集上的实验表明，信息密度可显著降低模型偏差，同时有效提升现有损失函数的整体性能。本研究为理解目标检测模型的广义偏差现象提供了新视角，为设计更公平的损失函数和训练策略提供了新工具。
英文摘要
In object detection, the number of instances is typically used to determine whether a dataset exhibits a long-tailed distribution, implicitly assuming that the model will perform poorly on categories with fewer instances. This assumption has led to extensive research on category bias in datasets with imbalanced instance numbers. However, even in datasets where instance numbers are relatively balanced, models still exhibit category bias, indicating that instance count alone cannot explain this phenomenon. In this work, we first introduce the concept and measurement of information density. We then observe a significant negative correlation between a category's information density and its accuracy, and we investigate how the training process impacts this relationship. Empirical studies suggest that information density imbalance may be a potential source of category bias. To preliminarily validate the potential of information density, we made simple improvements to three advanced object detection loss functions using this concept. Experiments on the Pascal VOC, COCO-LT, and LVIS datasets demonstrate that information density can significantly reduce model bias while effectively enhancing the overall performance of existing loss functions. This study provides a new perspective for understanding the generalized bias phenomenon in object detection models and offers new tools for designing fairer loss functions and training strategies.
