Learning Calibrated-Guidance for Object Detection in Aerial Images
Object detection is one of the most fundamental yet challenging research topics in the domain of computer vision. Recently, the study on this topic in aerial images has made tremendous progress. However, complex background and worse imaging quality are obvious problems in aerial object detection. Mo...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9735375/ |
| _version_ | 1818059609810665472 |
|---|---|
| author | Zongqi Wei Dong Liang Dong Zhang Liyan Zhang Qixiang Geng Mingqiang Wei Huiyu Zhou |
| author_facet | Zongqi Wei Dong Liang Dong Zhang Liyan Zhang Qixiang Geng Mingqiang Wei Huiyu Zhou |
| author_sort | Zongqi Wei |
| collection | DOAJ |
| description | Object detection is one of the most fundamental yet challenging research topics in the domain of computer vision. Recently, the study on this topic in aerial images has made tremendous progress. However, complex background and worse imaging quality are obvious problems in aerial object detection. Most state-of-the-art approaches tend to develop elaborate attention mechanisms for the space-time feature calibrations with arduous computational complexity, while surprisingly ignoring the importance of feature calibrations in channel-wise. In this work, we propose a simple yet effective calibrated-guidance (CG) scheme to enhance channel communications in a feature transformer fashion, which can adaptively determine the calibration weights for each channel based on the global feature affinity correlations. Specifically, for a given set of feature maps, CG first computes the feature similarity between each channel and the remaining channels as the intermediary calibration guidance. Then, rerepresenting each channel by aggregating all the channels weighted together via the guidance operation. Our CG is a general module that can be plugged into any deep neural networks, which is named as CG-Net. To demonstrate its effectiveness and efficiency, extensive experiments are carried out on both oriented object detection task and horizontal object detection task in aerial images. Experimental results on two challenging benchmarks (<italic>i</italic>.<italic>e</italic>., DOTA and HRSC2016) demonstrate that our CG-Net can achieve the new state-of-the-art performance in accuracy with a fair computational overhead. The source code has been open sourced at <uri>https://github.com/WeiZongqi/CG-Net</uri>. |
| first_indexed | 2024-12-10T13:19:15Z |
| format | Article |
| id | doaj.art-9505ae496d0b49f997ac1140e64ef674 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-10T13:19:15Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-9505ae496d0b49f997ac1140e64ef6742022-12-22T01:47:24ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352022-01-01152721273310.1109/JSTARS.2022.31589039735375Learning Calibrated-Guidance for Object Detection in Aerial ImagesZongqi Wei0https://orcid.org/0000-0001-6947-5280Dong Liang1https://orcid.org/0000-0003-2784-3449Dong Zhang2https://orcid.org/0000-0002-4543-2179Liyan Zhang3https://orcid.org/0000-0002-1549-3317Qixiang Geng4Mingqiang Wei5Huiyu Zhou6https://orcid.org/0000-0003-1634-9840College of Computer Science and Technology, MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Computer Science and Technology, MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaSchool of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, ChinaCollege of Computer Science and Technology, MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Computer Science and Technology, MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaCollege of Computer Science and Technology, MIIT Key Laboratory of Pattern Analysis and Machine Intelligence, Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing University of Aeronautics and Astronautics, Nanjing, ChinaSchool of Informatics, University of Leicester, Leicester, U.K.Object detection is one of the most fundamental yet challenging research topics in the domain of computer vision. Recently, the study on this topic in aerial images has made tremendous progress. However, complex background and worse imaging quality are obvious problems in aerial object detection. Most state-of-the-art approaches tend to develop elaborate attention mechanisms for the space-time feature calibrations with arduous computational complexity, while surprisingly ignoring the importance of feature calibrations in channel-wise. In this work, we propose a simple yet effective calibrated-guidance (CG) scheme to enhance channel communications in a feature transformer fashion, which can adaptively determine the calibration weights for each channel based on the global feature affinity correlations. Specifically, for a given set of feature maps, CG first computes the feature similarity between each channel and the remaining channels as the intermediary calibration guidance. Then, rerepresenting each channel by aggregating all the channels weighted together via the guidance operation. Our CG is a general module that can be plugged into any deep neural networks, which is named as CG-Net. To demonstrate its effectiveness and efficiency, extensive experiments are carried out on both oriented object detection task and horizontal object detection task in aerial images. Experimental results on two challenging benchmarks (<italic>i</italic>.<italic>e</italic>., DOTA and HRSC2016) demonstrate that our CG-Net can achieve the new state-of-the-art performance in accuracy with a fair computational overhead. The source code has been open sourced at <uri>https://github.com/WeiZongqi/CG-Net</uri>.https://ieeexplore.ieee.org/document/9735375/Aerial imageattention learningcalibrated-guidance (CG)deep learningobject detection |
| spellingShingle | Zongqi Wei Dong Liang Dong Zhang Liyan Zhang Qixiang Geng Mingqiang Wei Huiyu Zhou Learning Calibrated-Guidance for Object Detection in Aerial Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Aerial image attention learning calibrated-guidance (CG) deep learning object detection |
| title | Learning Calibrated-Guidance for Object Detection in Aerial Images |
| title_full | Learning Calibrated-Guidance for Object Detection in Aerial Images |
| title_fullStr | Learning Calibrated-Guidance for Object Detection in Aerial Images |
| title_full_unstemmed | Learning Calibrated-Guidance for Object Detection in Aerial Images |
| title_short | Learning Calibrated-Guidance for Object Detection in Aerial Images |
| title_sort | learning calibrated guidance for object detection in aerial images |
| topic | Aerial image attention learning calibrated-guidance (CG) deep learning object detection |
| url | https://ieeexplore.ieee.org/document/9735375/ |
| work_keys_str_mv | AT zongqiwei learningcalibratedguidanceforobjectdetectioninaerialimages AT dongliang learningcalibratedguidanceforobjectdetectioninaerialimages AT dongzhang learningcalibratedguidanceforobjectdetectioninaerialimages AT liyanzhang learningcalibratedguidanceforobjectdetectioninaerialimages AT qixianggeng learningcalibratedguidanceforobjectdetectioninaerialimages AT mingqiangwei learningcalibratedguidanceforobjectdetectioninaerialimages AT huiyuzhou learningcalibratedguidanceforobjectdetectioninaerialimages |