Monocular 3D Object Detection With Motion Feature Distillation

doi:10.1109/ACCESS.2023.3300708

CIOMP OpenIR

	Monocular 3D Object Detection With Motion Feature Distillation
	H. Hu; M. Li; M. Zhu; W. Gao; P. Liu and K.-L. Chan
	2023
发表期刊	IEEE Access
ISSN	21693536
卷号	11 页码:82933-82945
摘要	In the context of autonomous driving, environmental perception within a 360-degree field of view is extremely important. This can be achieved via the detection of three-dimensional (3D) objects in the surrounding scene with the inputs acquired by sensors such as LiDAR or RGB camera. The 3D perception generated is commonly represented as the bird's-eye-view (BEV) of the sensor. RGB camera has the advantages of low-cost and long-range acquisition. As the RGB images are two-dimensional (2D), the BEV generated from 2D images suffers from low accuracy due to limitations such as lack of temporal correlation. To address the problems, we propose a monocular 3D object detection method based on long short-term feature fusion and motion feature distillation. Long short-term temporal features are extracted with different feature map resolutions. The motion features and depth information are combined and encoded using an encoder based on the Transformer cross-correlation module, and further integrated into the BEV space of fused long short-term temporal features. Subsequently, a decoder with motion feature distillation is used to localize objects in 3D space. By combining BEV feature representations of different time steps, and supplemented with embedded motion features and depth information, our proposed method significantly improves the accuracy of monocular 3D object detection as demonstrated from experimental results obtained on nuScenes dataset. Our proposed method outperforms state-of-the-art methods, in particular the previous best art by 6.7% on mAP, and 8.3% on mATE. © 2013 IEEE.
DOI	10.1109/ACCESS.2023.3300708
URL	查看原文
收录类别	sci ; ei
引用统计
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/67530
专题	中国科学院长春光学精密机械与物理研究所
推荐引用方式 GB/T 7714	H. Hu,M. Li,M. Zhu,et al. Monocular 3D Object Detection With Motion Feature Distillation[J]. IEEE Access,2023,11:82933-82945.
APA	H. Hu,M. Li,M. Zhu,W. Gao,&P. Liu and K.-L. Chan.(2023).Monocular 3D Object Detection With Motion Feature Distillation.IEEE Access,11,82933-82945.
MLA	H. Hu,et al."Monocular 3D Object Detection With Motion Feature Distillation".IEEE Access 11(2023):82933-82945.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Monocular 3D Object （1431KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载