Induced subgraph searching for geometric model fitting

Fan Xiao; Guobao Xiao; Yan Yan; Xing Wang; Hanzi Wang

doi:10.1117/12.2296331

15 November 2017 Induced subgraph searching for geometric model fitting

Fan Xiao, Guobao Xiao, Yan Yan, Xing Wang, Hanzi Wang

Proceedings Volume 10605, LIDAR Imaging Detection and Target Recognition 2017; 106054B (2017) https://doi.org/10.1117/12.2296331
Event: LIDAR Imaging Detection and Target Recognition 2017, 2017, Changchun, China

Abstract

In this paper, we propose a novel model fitting method based on graphs to fit and segment multiple-structure data. In the graph constructed on data, each model instance is represented as an induced subgraph. Following the idea of pursuing the maximum consensus, the multiple geometric model fitting problem is formulated as searching for a set of induced subgraphs including the maximum union set of vertices. After the generation and refinement of the induced subgraphs that represent the model hypotheses, the searching process is conducted on the “qualified” subgraphs. Multiple model instances can be simultaneously estimated by solving a converted problem. Then, we introduce the energy evaluation function to determine the number of model instances in data. The proposed method is able to effectively estimate the number and the parameters of model instances in data severely corrupted by outliers and noises. Experimental results on synthetic data and real images validate the favorable performance of the proposed method compared with several state-of-the-art fitting methods.

Citation Download Citation

Fan Xiao, Guobao Xiao, Yan Yan, Xing Wang, and Hanzi Wang "Induced subgraph searching for geometric model fitting", Proc. SPIE 10605, LIDAR Imaging Detection and Target Recognition 2017, 106054B (15 November 2017); https://doi.org/10.1117/12.2296331

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available