HDR-Net-Fusion:Real-time 3D dynamic scene reconstruction with a hierarchical deep reinforcement network

作     者：Hao-Xuan Song Jiahui Huang Yan-Pei Cao Tai-Jiang Mu Hao-Xuan Song;Jiahui Huang;Yan-Pei Cao;Tai-Jiang Mu

作者机构：BNRistDepartment of Computer Science and TechnologyTsinghua UniversityBeiing 100084China Kuaishou Technology Co.Ltd.Beijing 100085China 

出 版 物：《Computational Visual Media》 (计算可视媒体（英文版）)

年 卷 期：2021年第7卷第4期

页      面：419-435页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0835[工学-软件工程] 0802[工学-机械工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was supported by the National Natural Science Foundation of China(Grant Nos.61902210 and 61521002) 

主　　题：dynamic 3D scene reconstruction deep reinforcement learning point cloud completion deep neural networks 

摘      要：Reconstructing dynamic scenes with commodity depth cameras has many applications in computer graphics,computer vision,and ***,due to the presence of noise and erroneous observations from data capturing devices and the inherently ill-posed nature of non-rigid registration with insufficient information,traditional approaches often produce low-quality geometry with holes,bumps,and *** propose a novel 3D dynamic reconstruction system,named HDR-Net-Fusion,which learns to simultaneously reconstruct and refine the geometry on the fly with a sparse embedded deformation graph of surfels,using a hierarchical deep reinforcement(HDR)*** latter comprises two parts:a global HDR-Net which rapidly detects local regions with large geometric errors,and a local HDR-Net serving as a local patch refinement operator to promptly complete and enhance such *** the global HDR-Net is formulated as a novel reinforcement learning problem to implicitly learn the region selection strategy with the goal of improving the overall reconstruction *** applicability and efficiency of our approach are demonstrated using a large-scale dynamic reconstruction *** method can reconstruct geometry with higher quality than traditional methods.