中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Progressive Depth Decoupling and Modulating for Flexible Depth Completion

文献类型：期刊论文

作者Yang, Zhiwen1; Zhang, Jiehua2; Li, Liang3; Yan, Chenggang4; Sun, Yaoqi5,6; Yin, Haibing5,6
刊名IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
出版日期2024
卷号73页码:16
关键词Accuracy Decoding Transformers Three-dimensional displays Task analysis Estimation Research and development Adaptive depth modulating depth completion depth discretization incremental depth decoupling
ISSN号0018-9456
DOI10.1109/TIM.2024.3420352
英文摘要Image-guided depth completion aims at generating a dense depth map from sparse light detection and ranging (LiDAR) data and the corresponding RGB image, which is crucial for applications that require 3-D scene perception, such as augmented reality and human-computer interaction. Recent methods have shown promising performance by reformulating it as a classification problem with two subtasks: depth discretization and probability prediction. They divide the depth range into several discrete depth values as depth categories, serving as priors for scene depth distributions. However, previous depth discretization methods are easy to be impacted by depth distribution variations across different scenes, resulting in suboptimal scene depth distribution priors. To address the above problem, we propose a progressive depth decoupling and modulating network, which incrementally decouples the depth range into bins and adaptively generates multiscale dense depth maps in multiple stages. Specifically, we first design a bins initializing module (BIM) to construct the seed bins by exploring the depth distribution information within a sparse depth map, adapting variations of depth distribution. Then, we devise an incremental depth decoupling branch to progressively refine the depth distribution information from global to local. Meanwhile, an adaptive depth modulating branch is developed to progressively improve the probability representation from coarse-grained to fine-grained. Also, the bidirectional information interactions are proposed to strengthen the information interaction between those two branches (subtasks) for promoting information complementation in each branch. Furthermore, we introduce a multiscale supervision mechanism to learn the depth distribution information in latent features and enhance the adaptation capability across different scenes. Experimental results on public datasets demonstrate that our method outperforms the state-of-the-art (SOTA) methods. We will release the source codes and pretrained models.
资助项目National Key Research and Development Program of China[2023YFB4502800] ; National Key Research and Development Program of China[2023YFB4502803] ; National Key Research and Development Program of China[2020YFB1406604] ; National Natural Science Foundation of China[61931008] ; National Natural Science Foundation of China[U21B2024] ; National Natural Science Foundation of China[62071415] ; National Natural Science Foundation of China[62322211] ; National Natural Science Foundation of China[62336008] ; Zhejiang Provincial Natural Science Foundation of China[LDT23F01011F01] ; Zhejiang Provincial Natural Science Foundation of China[LDT23F01015F01] ; Zhejiang Provincial Natural Science Foundation of China[LDT23F01014F01] ; Pioneer and Leading Goose Research and Development Program of Zhejiang Province[2022C01068] ; Key Research and Development Plan Project of Zhejiang Province[2024C01023]
WOS研究方向Engineering ; Instruments & Instrumentation
语种英语
WOS记录号WOS:001282362800004
出版者IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
源URL[http://119.78.100.204/handle/2XEOYT63/39695]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Li, Liang
作者单位1.Hangzhou Dianzi Univ, Sch Automation, Hangzhou 310018, Peoples R China
2.Xi An Jiao Tong Univ, Sch Software Engn, Xian 614202, Peoples R China
3.Chinese Acad Sci, Inst Comp Technol, Beijing 100089, Peoples R China
4.Hangzhou Dianzi Univ, Sch Commun Engn, Hangzhou 310018, Peoples R China
5.Hangzhou Dianzi Univ, Sch Commun Engn, Hangzhou 310018, Peoples R China
6.Hangzhou Dianzi Univ, Lishui Inst, Hangzhou 310018, Peoples R China
推荐引用方式
GB/T 7714		 Yang, Zhiwen,Zhang, Jiehua,Li, Liang,et al. Progressive Depth Decoupling and Modulating for Flexible Depth Completion[J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,2024,73:16.
APA Yang, Zhiwen,Zhang, Jiehua,Li, Liang,Yan, Chenggang,Sun, Yaoqi,&Yin, Haibing.(2024).Progressive Depth Decoupling and Modulating for Flexible Depth Completion.IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,73,16.
MLA Yang, Zhiwen,et al."Progressive Depth Decoupling and Modulating for Flexible Depth Completion".IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 73(2024):16.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
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。