一种在密集场景中检测物体的方法

IF 16.4 1区化学 Q1 CHEMISTRY, MULTIDISCIPLINARY

Accounts of Chemical Research Pub Date : 2022-01-01 DOI:10.1515/comp-2022-0231

Chuanyun Xu, Yueping Zheng, Yang Zhang, Gang Li, Ying Wang

{"title":"一种在密集场景中检测物体的方法","authors":"Chuanyun Xu, Yueping Zheng, Yang Zhang, Gang Li, Ying Wang","doi":"10.1515/comp-2022-0231","DOIUrl":null,"url":null,"abstract":"Abstract Recent object detectors have achieved excellent performance in accuracy and speed. Even with such impressive results, the most advanced detectors are challenging in dense scenes. In this article, we analyze and find the reasons for the decrease in detection accuracy in dense scenes. We started our work in terms of region proposal and location loss. We found that low-quality proposal regions during the training process are the main factors affecting detection accuracy. To prove our research, we established and trained a dense detection model based on Cascade R-CNN. The model achieves an accuracy of mAP 0.413 on the SKU-110K sub-dataset. Our results show that improving the quality of recommended regions can effectively improve the detection accuracy in dense scenes.","PeriodicalId":1,"journal":{"name":"Accounts of Chemical Research","volume":null,"pages":null},"PeriodicalIF":16.4000,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"A method for detecting objects in dense scenes\",\"authors\":\"Chuanyun Xu, Yueping Zheng, Yang Zhang, Gang Li, Ying Wang\",\"doi\":\"10.1515/comp-2022-0231\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract Recent object detectors have achieved excellent performance in accuracy and speed. Even with such impressive results, the most advanced detectors are challenging in dense scenes. In this article, we analyze and find the reasons for the decrease in detection accuracy in dense scenes. We started our work in terms of region proposal and location loss. We found that low-quality proposal regions during the training process are the main factors affecting detection accuracy. To prove our research, we established and trained a dense detection model based on Cascade R-CNN. The model achieves an accuracy of mAP 0.413 on the SKU-110K sub-dataset. Our results show that improving the quality of recommended regions can effectively improve the detection accuracy in dense scenes.\",\"PeriodicalId\":1,\"journal\":{\"name\":\"Accounts of Chemical Research\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":16.4000,\"publicationDate\":\"2022-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Accounts of Chemical Research\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1515/comp-2022-0231\",\"RegionNum\":1,\"RegionCategory\":\"化学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"CHEMISTRY, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Accounts of Chemical Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1515/comp-2022-0231","RegionNum":1,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 2

摘要

摘要近年来，物体探测器在精度和速度方面都取得了优异的性能。即使有如此令人印象深刻的结果，最先进的探测器在密集的场景中也是具有挑战性的。在这篇文章中，我们分析并找出了在密集场景中检测精度下降的原因。我们从区域建议和位置损失开始了我们的工作。我们发现，训练过程中的低质量建议区域是影响检测准确性的主要因素。为了证明我们的研究，我们建立并训练了一个基于级联R-CNN的密集检测模型。该模型在SKU-110K子数据集上实现了mAP 0.413的精度。我们的结果表明，提高推荐区域的质量可以有效地提高密集场景中的检测精度。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A method for detecting objects in dense scenes

Abstract Recent object detectors have achieved excellent performance in accuracy and speed. Even with such impressive results, the most advanced detectors are challenging in dense scenes. In this article, we analyze and find the reasons for the decrease in detection accuracy in dense scenes. We started our work in terms of region proposal and location loss. We found that low-quality proposal regions during the training process are the main factors affecting detection accuracy. To prove our research, we established and trained a dense detection model based on Cascade R-CNN. The model achieves an accuracy of mAP 0.413 on the SKU-110K sub-dataset. Our results show that improving the quality of recommended regions can effectively improve the detection accuracy in dense scenes.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Accounts of Chemical Research 化学-化学综合

CiteScore

31.40

自引率

1.10%

发文量

312

审稿时长

2 months

期刊介绍： Accounts of Chemical Research presents short, concise and critical articles offering easy-to-read overviews of basic research and applications in all areas of chemistry and biochemistry. These short reviews focus on research from the author’s own laboratory and are designed to teach the reader about a research project. In addition, Accounts of Chemical Research publishes commentaries that give an informed opinion on a current research problem. Special Issues online are devoted to a single topic of unusual activity and significance. Accounts of Chemical Research replaces the traditional article abstract with an article "Conspectus." These entries synopsize the research affording the reader a closer look at the content and significance of an article. Through this provision of a more detailed description of the article contents, the Conspectus enhances the article's discoverability by search engines and the exposure for the research.