聚合多维注意力特征的单目深度估计方法

doi:10.3969/j.issn.1006-2475.2023.06.013

计算机与现代化 ›› 2023, Vol. 0 ›› Issue (06): 76-81.doi: 10.3969/j.issn.1006-2475.2023.06.013

聚合多维注意力特征的单目深度估计方法

刘甲甲¹, 胡旭欣², 余萍²

1.深圳供电规划设计院有限公司,广东深圳518000;
2.华北电力大学（保定)电气与电子工程学院,河北保定 071000

收稿日期:2022-06-06 修回日期:2022-07-17 出版日期:2023-06-28 发布日期:2023-06-28
通讯作者: 余萍（1963—）,女,副教授,本科,研究方向：图像处理,无线通信,E-mail： well_yp@sina.com。
作者简介:刘甲甲（1985—）,男,安徽淮南人,工程师,本科,研究方向：计算机应用,E-mail： 181509351@qq.com; 胡旭欣（1994—）,女,硕士,研究方向：深度学习,E-mail： 446844661@qq.com。
基金资助:
中国南方电网公司深圳供电局有限公司科技项目（090000KK52180035）

Monocular Depth Estimation Method by Aggregating Multi-dimensional Attention Features

LIU Jia-jia¹, HU Xu-xin², YU Ping²

1. Shenzhen Power Supply Planning and Design Institute Co., Ltd, Shenzhen 518000, China;
2. School of Electrical and Electronic Engineering, North China Electric Power University, Baoding 071000, China

Received:2022-06-06 Revised:2022-07-17 Online:2023-06-28 Published:2023-06-28

摘要/Abstract

摘要： 为了提升单目深度估计网络的预测精度,本文深入研究多维注意力机制对单目深度估计网络的影响与作用,并设计一种优化后的通道和空间注意力模块。在基于局部平面指导层的卷积神经网络框架上,通过设计模块的放置方法,构建可以充分激发多维注意力机制有效性的网络结构。结合以上2点改进措施,得到一种高性能的单目深度估计网络——聚合通道和空间注意力特征的单目深度估计网络。在KITTI深度数据集和NYU Depth V2数据集中,通过实验分别验证优化模块的有效性和聚合网络的优秀性能。相比于基于局部平面指导层的卷积神经网络,聚合网络对于图像的总体特征具有更好的处理能力,预测的深度信息更加精准,网络的多个评价指标均有不同幅度的提升。同时,聚合网络生成的深度图也展现出了更多的物体轮廓和细节信息。

关键词: 单目深度估计, 卷积神经网络, 通道注意力, 空间注意力

Abstract: This study is outlined to improve the precision for predicting monocular depth estimation networks and provides an in-depth analysis of the effects of multidimensional attention mechanisms on monocular depth estimation networks. The conclusions and observations are used to design a set of optimized channel and space attention blocks. Considering the convolutional neural network framework obtained based on the local plane guidance layer, a new network structure is created to fully activate the multidimensional attention mechanism through a method that is based on placing different design blocks. Furthermore, in combination with the above two measures for improvement, this study proposes a high-performance monocular depth estimation network that integrates channel and space attention features. On the KITTI Depth dataset and an NYU Depth V2 dataset, the outcomes of this study prove the effectiveness of the optimized blocks and the satisfactory performance of the proposed network through experiments. Compared with the convolutional neural network based on the local plane guidance layer, the proposed network is better in processing the overall features of images and more accurate in predicting depth information with several metrics for network evaluation improved to different degrees. The depth maps generated by the proposed network also demonstrated more data associated with the contours and details of objects.

Key words: monocular depth estimation, convolutional neural network, channel attention, spatial attention

中图分类号:

TP391

刘甲甲, 胡旭欣, 余萍. 聚合多维注意力特征的单目深度估计方法[J]. 计算机与现代化, 2023, 0(06): 76-81.

LIU Jia-jia, HU Xu-xin, YU Ping. Monocular Depth Estimation Method by Aggregating Multi-dimensional Attention Features[J]. Computer and Modernization, 2023, 0(06): 76-81.

参考文献 30

[1]	SAXENA A, CHUNG S H, NG A Y.Learning depth from single monocular images[C]// Proceedings of the 18th International Conference on Neural Information Processing Systems. 2005:1161-1168.
[2]	EIGEN D, PUHRSCH C, FERGUS R.Depth map prediction from a single image using a multi-scale deep network[C]// Proceedings of 2015 IEEE International Conference on Computer Vision (ICCV). 2015:2650-2658.
[3]	王泉德,张松涛. 基于多尺度特征融合的单目图像深度估计[J]. 华中科技大学学报(自然科学版), 2020,48(5):7-12.
[4]	贾瑞明,李阳,李彤,等. 多层级特征融合结构的单目图像深度估计网络[J]. 计算机工程, 2020,46(12):207-214.
[5]	LI B, SHEN C, DAI Y, et al.Depth and surface normal estimation from monocular images using regression on deep features and hierarchical CRFs[C]// 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2015:1119-1127.
[6]	LAINA I, RUPPRECHT C, BELAGIANNIS V, et al.Deeper depth prediction with fully convolutional residual networks[C]// 2016 4th International Conference on 3D Vision (3DV). 2016:239-248.
[7]	袁建中,周武杰,潘婷,等. 基于深度卷积神经网络的道路场景深度估计[J]. 激光与光电子学进展, 2019,56(8):179-187.
[8]	GAN Y, XU X, SUN W, et al.Monocular depth estimation with affinity, vertical pooling, and label enhancement[C]// Computer Vision - ECCV 2018. 2018:232-247.
[9]	YIN W, LIU Y, SHEN C, et al.Enforcing geometric constraints of virtual normal for depth prediction[C]// 2019 IEEE/CVF International Conference on Computer Vision (ICCV). 2019:5683-5692.
[10]	RANFTL R, LASINGER K, HAFNER D, et al.Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022,44(3):1623-1637.
[11]	FU H, GONG M, WANG C, et al.Deep ordinal regression network for monocular depth estimation[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018:2002-2011.
[12]	ALHASHIM I, WONKA P. High quality monocular depth estimation via transfer learning[J]. arXiv preprint arXiv:1812.11941, 2018.
[13]	LEE J H, HAN M, KO D W, et al.From big to small: Multi-scale local planar guidance for monocular depth estimation[J]. arXiv preprint arXiv:1907.10326, 2019.
[14]	QIAO S, ZHU Y, ADAM H, et al.ViP-DeepLab: Learning visual perception with depth-aware video panoptic segmentation[C]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021:3996-4007.
[15]	廖志伟,金兢,张超凡,等. 基于分层压缩激励的ASPP网络单目深度估计[J]. 图学学报, 2022,43(2):214-222.
[16]	王泉德,王奇坤,程凯,等. 强化边缘的单目图像深度估计[J]. 华中科技大学学报(自然科学版), 2022,50(3):36-42.
[17]	刘杰平,温竣文,梁亚玲. 基于多尺度注意力导向网络的单目图像深度估计[J]. 华南理工大学学报(自然科学版), 2020,48(12):52-62.
[18]	WANG Q, WU B, ZHU P, et al.ECA-Net: Efficient channel attention for deep convolutional neural networks[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020:11531-11539.
[19]	AICH S, VIANNEY J M U, ISLAM M A, et al. Bidirectional attention network for monocular depth estimation[C]// 2021 IEEE International Conference on Robotics and Automation (ICRA). 2021:11746-11752.
[20]	LEE S, LEE J, KIM B, et al.Patch-wise attention network for monocular depth estimation[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2021,35(3):1873-1881.
[21]	CHEN W, ZHAO F, YANG D W, et al.Single-image depth perception in the wild[C]// 2016 30th Conference on Neural Information Processing System. 2016:730-738.
[22]	曾一鸣. 利用稀疏点云偏序关系的半监督单目图像深度估计[J]. 计算机辅助设计与图形学学报, 2019,31(11):2038-2046.
[23]	邹承明,胡佑璞. 引入生成对抗网络的室外场景单目深度估计[J]. 计算机工程与应用, 2021,57(6):176-183.
[24]	GARG R, BG V K, CARNEIRO G, et al.Unsupervised CNN for single view depth estimation: Geometry to the rescue[C]// Computer Vision - ECCV 2016. 2016:740-756.
[25]	GODARD C, AODHA O M, BROSTOW G J.Unsupervised monocular depth estimation with left-right consistency[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017:6602-6611.
[26]	刘香凝,赵洋,王荣刚. 基于自注意力机制的多阶段无监督单目深度估计网络[J]. 信号处理, 2020,36(9):1450-1456.
[27]	CHEN L, PAPANDREOU G, KOKKINOS I, et al.DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018,40(4):834-848.
[28]	WOO S, PARK J, LEE J, et al.CBAM: Convolutional block attention module[C]// Computer Vision - ECCV 2018. 2018:3-19.
[29]	XIE S, GIRSHICK R, DOLLÄR P, et al. Aggregated residual transformations for deep neural networks[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017:5987-5995.
[30]	HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017:4700-4708.

聚合多维注意力特征的单目深度估计方法

Monocular Depth Estimation Method by Aggregating Multi-dimensional Attention Features

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献 30

相关文章 15

编辑推荐

Metrics

本文评价

[1]	周成诚, 曾庆军, 杨康, 胡家铭, 韩春伟. 基于高效通道注意力模块的运动想象脑电识别[J]. 计算机与现代化, 2023, 0(12): 19-23.
[2]	张浩洋, 尹梓名, 乐珺怡, 沈达聪, 束翌俊, 杨自逸, 孔祥勇, 龚伟. 3D-SPRNet: 一种基于并行解码器和双注意力机制的胆囊癌分割模型[J]. 计算机与现代化, 2023, 0(12): 59-66.
[3]	邢世帅, 刘丹凤, 王立国, 潘月涛, 孟灵鸿, 岳晓晗. 基于空间注意力残差网络的图像超分辨率重建模型[J]. 计算机与现代化, 2023, 0(10): 45-52.
[4]	刘付琪, 张达, 宋建华, 王海东. 基于CNN-BiLSTM的液压系统故障诊断[J]. 计算机与现代化, 2023, 0(09): 10-19.
[5]	吴甜, 刘海华, 童顺延. 基于深度反馈的卷积神经网络的图像分类[J]. 计算机与现代化, 2023, 0(09): 82-86.
[6]	孙子雨, 任燃, 魏曦哲. 基于DTW-TCN的股票分类及预测研究[J]. 计算机与现代化, 2023, 0(08): 31-37.
[7]	江蕾, 唐建, 杨超越, 吕婷婷. 基于CWGAN-GP与CNN的轴承故障诊断方法[J]. 计算机与现代化, 2023, 0(07): 1-6.
[8]	许叶彤, 耿信哲, 赵伟强, 张月, 宁海龙, 雷涛. 基于CNN-Transformer混合结构的遥感影像变化检测模型[J]. 计算机与现代化, 2023, 0(07): 79-85.
[9]	朱剑波, 葛明锋, 董文飞. 基于改进EfficientNet的阿尔兹海默症图像分类[J]. 计算机与现代化, 2023, 0(06): 56-61.
[10]	刘静, 陈金广. 基于通道注意力和Transformer的图像标题生成方法[J]. 计算机与现代化, 2023, 0(05): 8-12.
[11]	王欣怡, 尹四清, 洪军. 融合注意力机制的非对称深度监督哈希[J]. 计算机与现代化, 2023, 0(05): 26-31.
[12]	苏金库, 桂智明. 基于时空特征的短时出租车流量预测[J]. 计算机与现代化, 2023, 0(05): 32-38.
[13]	王娟, 李传庚, 张卿源, 夏承遗. 基于Multiscale-Net的膝关节半月板分割方法[J]. 计算机与现代化, 2023, 0(05): 111-116.
[14]	王磊, 张晓东, 戴欢. 基于1D-CNN-LSTM注意力网络的抽油机井故障诊断[J]. 计算机与现代化, 2023, 0(04): 1-6.
[15]	徐涯昕, 何泽恩, 徐绪堪. 基于CNN-BiLSTM网络的数控机床故障文本自动分类[J]. 计算机与现代化, 2023, 0(04): 7-14.