Unsupervised learning of depth estimation from imperfect rectified stereo laparoscopic images
文献类型:期刊论文
作者 | Luo HL(罗火灵)5,6; Wang, Congcong7,8; Duan, Xingguang1; Liu H(刘浩)2; Wang, Ping4; Hu, QM(胡庆茂)5,6; Jia FC(贾富仓)3,5,6 |
刊名 | Computers in Biology and Medicine |
出版日期 | 2022 |
卷号 | 140页码:1-12 |
ISSN号 | 0010-4825 |
关键词 | Unsupervised learning Depth estimation Stereo matching Laparoscopic image Imperfect rectified stereo images |
产权排序 | 6 |
英文摘要 | Background: Learning-based methods have achieved remarkable performances on depth estimation. However, the premise of most self-learning and unsupervised learning methods is built on rigorous, geometrically-aligned stereo rectification. The performances of these methods degrade when the rectification is not accurate. Therefore, we explore an approach for unsupervised depth estimation from stereo images that can handle imperfect camera parameters. Methods: We propose an unsupervised deep convolutional network that takes rectified stereo image pairs as input and outputs corresponding dense disparity maps. First, a new vertical correction module is designed for predicting a correction map to compensate for the imperfect geometry alignment. Second, the left and right images, which are reconstructed based on the input image pair and corresponding disparities as well as the vertical correction maps, are regarded as the outputs of the generative term of the generative adversarial network (GAN). Then, the discriminator term of the GAN is used to distinguish the reconstructed images from the original inputs to force the generator to output increasingly realistic images. In addition, a residual mask is introduced to exclude pixels that conflict with the appearance of the original image in the loss calculation. Results: The proposed model is validated on the publicly available Stereo Correspondence and Reconstruction of Endoscopic Data (SCARED) dataset and the average MAE is 3.054 mm. Conclusion: Our model can effectively handle imperfect rectified stereo images for depth estimation. |
WOS关键词 | MINIMALLY INVASIVE SURGERY ; SURFACE RECONSTRUCTION ; 3D RECONSTRUCTION ; DENSE |
资助项目 | NSFC[62172401] ; NSFC[12026602] ; NSFC[62102285] ; National Key R&D Program, China[2019YFC0118100] ; National Key R&D Program, China[2017YFC0110903] ; Key-Area Research and Development Program of Guangdong Province, China[2020B010165004] ; Shenzhen Key Basic Science Program[JCYJ20180507182437217] ; Shenzhen Key Laboratory Program[ZDSYS201707271637577] |
WOS研究方向 | Life Sciences & Biomedicine - Other Topics ; Computer Science ; Engineering ; Mathematical & Computational Biology |
语种 | 英语 |
WOS记录号 | WOS:000731856100003 |
资助机构 | NSFC Grant Program (62172401, 12026602 and 62102285) ; National Key R&D Program, China (Nos. 2019YFC0118100 and 2017YFC0110903) ; Key-Area Research and Development Program of Guangdong Province, China (No. 2020B010165004) ; Shenzhen Key Basic Science Program (No. JCYJ20180507182437217) ; Shenzhen Key Laboratory Program (ZDSYS201707271637577) |
源URL | [http://ir.sia.cn/handle/173321/30095] |
专题 | 沈阳自动化研究所_机器人学研究室 |
通讯作者 | Jia FC(贾富仓) |
作者单位 | 1.Advanced Innovation Centre for Intelligent Robots & Systems, Beijing Institute of Technology, Beijing, China 2.State Key Lab for Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China 3.Pazhou Lab, Guangzhou, China 4.Department of Hepatobiliary Surgery, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China 5.Research Lab for Medical Imaging and Digital Surgery, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China 6.Shenzhen College of Advanced Technology, University of Chinese Academy of Sciences, Shenzhen, China 7.School of Computer Science and Engineering, Tianjin University of Technology, Tianjin, China 8.Department of Computer Science, Norwegian University of Science and Technology, Trondheim, Norway |
推荐引用方式 GB/T 7714 | Luo HL,Wang, Congcong,Duan, Xingguang,et al. Unsupervised learning of depth estimation from imperfect rectified stereo laparoscopic images[J]. Computers in Biology and Medicine,2022,140:1-12. |
APA | Luo HL.,Wang, Congcong.,Duan, Xingguang.,Liu H.,Wang, Ping.,...&Jia FC.(2022).Unsupervised learning of depth estimation from imperfect rectified stereo laparoscopic images.Computers in Biology and Medicine,140,1-12. |
MLA | Luo HL,et al."Unsupervised learning of depth estimation from imperfect rectified stereo laparoscopic images".Computers in Biology and Medicine 140(2022):1-12. |
入库方式: OAI收割
来源:沈阳自动化研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。