papers AI Learner
The Github is limit! Click to go to the new site.

Devon: Deformable Volume Network for Learning Optical Flow

2019-03-04
Yao Lu, Jack Valmadre, Heng Wang, Juho Kannala, Mehrtash Harandi, Philip H. S. Torr

Abstract

State-of-the-art neural network models estimate large displacement optical flow in multi-resolution and use warping to propagate the estimation between two resolutions. Despite their impressive results, it is known that there are two problems with the approach. First, the multi-resolution estimation of optical flow fails in situations where small objects move fast. Second, warping creates artifacts when occlusion or dis-occlusion happens. In this paper, we propose a new neural network module, Deformable Cost Volume, which alleviates the two problems. Based on this module, we designed the Deformable Volume Network (Devon) which can estimate multi-scale optical flow in a single high resolution. Experiments show Devon is more suitable in handling small objects moving fast and achieves comparable results to the state-of-the-art methods in public benchmarks.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1802.07351

PDF

http://arxiv.org/pdf/1802.07351


Similar Posts

Comments