Affiliations: [a] School of Electrical Engineering, Southwest Jiaotong University, Chengdu, Sichuan, China | [b] Institute of Artificial Intelligence, School of Computer Science and Informatics, De Montfort University, Leicester, UK | [c] Control Engineering College, Chengdu University of Information Technology, Chengdu, Sichuan, China | [d] Institute of Real Time Learning Systems, University of Siegen, Siegen, Germany
Correspondence:
[*]
Corresponding author: Gexiang Zhang, School of Electrical Engineering, Southwest Jiaotong University, Chengdu, Sichuan 610031, China. E-mail: [email protected].
Abstract: An artificial compound eye (ACE) is a bio-inspired vision sensor which mimics a natural compound eye (typical of insects). This artificial eye is able to visualize large fields of the outside world through multi-aperture. Due to its functioning, the ACE is subject to optical flow, that is an apparent motion of the object visualized by the eye. This paper proposes a method to estimate the optical flow based on capturing multiple images (multi-aperture). In this method, based on descriptors-based initial optical flows, a unified global energy function is presented to incorporate the information of multi-aperture and simultaneously recover the optical flows of multi-aperture. The energy function imposes a compound flow fields consistency assumption along with the brightness constancy and piecewise smoothness assumptions. This formula efficiently binds the flow field in time and space, and further enables view-consistent optical flow estimation. Experimental results on real and synthetic data demonstrate that the proposed method recovers view-consistent optical flows crossed multi-aperture and performs better than other optical flow methods on the multi-aperture images.
