Real-time Object Detection with Attention Mask

Haixin Wang, Xue Bai, Qiongzhi Wu

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

With the development of deep convolutional neural network, the performance of object detection is obviously improved. However, there are still some challenges such as small size and occlusion. In this paper, we present a novel detector named Attention Mask Detector (AMDet). Our motivation is using mask to enhance foreground features and suppress background ones. The mask is produced by an attention branch which is supervised by weak segmentation ground-truth. This weak segmentation ground-truth is generated by bounding box without extra annotations. Our method is based on one-stage detector. We do experiments on both PASCAL VOC and MS COCO datasets and have a result comparison with other one-stage detectors.

源语言英语
主期刊名ICSIDP 2019 - IEEE International Conference on Signal, Information and Data Processing 2019
出版商Institute of Electrical and Electronics Engineers Inc.
ISBN(电子版)9781728123455
DOI
出版状态已出版 - 12月 2019
活动2019 IEEE International Conference on Signal, Information and Data Processing, ICSIDP 2019 - Chongqing, 中国
期限: 11 12月 201913 12月 2019

出版系列

姓名ICSIDP 2019 - IEEE International Conference on Signal, Information and Data Processing 2019

会议

会议2019 IEEE International Conference on Signal, Information and Data Processing, ICSIDP 2019
国家/地区中国
Chongqing
时期11/12/1913/12/19

指纹

探究 'Real-time Object Detection with Attention Mask' 的科研主题。它们共同构成独一无二的指纹。

引用此