A Deep Learning Driven Active Framework for Segmentation of Large 3D Shape Collections

被引:4
作者
George, David [1 ]
Xie, Xianghua [1 ]
Lai, Yukun [2 ]
Tam, Gary K. L. [1 ]
机构
[1] Swansea Univ, Dept Comp Sci, Swansea, W Glam, Wales
[2] Cardiff Univ, Sch Comp Sci & Informat, Cardiff, Wales
基金
英国工程与自然科学研究理事会;
关键词
Shape segmentation; Active learning; Shape collections; User interaction; UNSUPERVISED CO-SEGMENTATION; OBJECT RECOGNITION; MACHINE;
D O I
10.1016/j.cad.2021.103179
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
High-level shape understanding and technique evaluation on large repositories of 3D shapes often benefit from additional information known about the shapes. One example of such information is the semantic segmentation of a shape into functional or meaningful parts. Generating accurate segmentations with meaningful segment boundaries is, however, a costly process, typically requiring large amounts of user time to achieve high-quality results. In this paper we propose an active learning framework for large dataset segmentation, which iteratively provides the user with new predictions by training new models based on already segmented shapes. Our proposed pipeline consists of three components. First, we propose a fast and accurate feature-based deep learning model to provide dataset-wide segmentation predictions. Second, we develop an information theory measure to estimate the prediction quality and for ordering subsequent fast and meaningful shape selection. Our experiments show that such suggestive ordering helps to reduce users' time and effort, produce high-quality predictions, and construct a model that generalizes well. Lastly, we provide interactive segmentation refinement tools, helping the user quickly correct any prediction errors. We show that our framework is more accurate and in general more efficient than the state-of-the-art for large dataset segmentation, while also providing consistent segment boundaries. (c) 2021 Elsevier Ltd. All rights reserved.
引用
收藏
页数:15
相关论文
共 56 条
[1]  
[Anonymous], 2011, P ACM SIGGRAPH ASIA
[2]   Hierarchical mesh segmentation based on fitting primitives [J].
Attene, M ;
Falcidieno, B ;
Spagnuolo, M .
VISUAL COMPUTER, 2006, 22 (03) :181-193
[3]   Shape matching and object recognition using shape contexts [J].
Belongie, S ;
Malik, J ;
Puzicha, J .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2002, 24 (04) :509-522
[4]  
Ben-Chen M, 2008, 3D OBJ RETRIEVAL, P1
[5]   Learning Boundary Edges for 3D-Mesh Segmentation [J].
Benhabiles, Halim ;
Lavoue, Guillaume ;
Vandeborre, Jean-Philippe ;
Daoudi, Mohamed .
COMPUTER GRAPHICS FORUM, 2011, 30 (08) :2170-2182
[6]   Fast approximate energy minimization via graph cuts [J].
Boykov, Y ;
Veksler, O ;
Zabih, R .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2001, 23 (11) :1222-1239
[7]   The Ignorant Led by the Blind: A Hybrid Human-Machine Vision System for Fine-Grained Categorization [J].
Branson, Steve ;
Van Horn, Grant ;
Wah, Catherine ;
Perona, Pietro ;
Belongie, Serge .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2014, 108 (1-2) :3-29
[8]  
Branson S, 2011, IEEE I CONF COMP VIS, P1832, DOI 10.1109/ICCV.2011.6126450
[9]  
Chang Angel X, 2015, Technical Report
[10]   On visual similarity based 3D model retrieval [J].
Chen, DY ;
Tian, XP ;
Shen, YT ;
Ming, OY .
COMPUTER GRAPHICS FORUM, 2003, 22 (03) :223-232