LiteFlowNet: A Lightweight Convolutional Neural Network for Optical Flow Estimation

• Published in: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition pp. 8981 - 8989
• Main Authors: Hui, Tak-Wai, Tang, Xiaoou, Loy, Chen Change
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2018
• Subjects: Adaptive optics; Convolution; Estimation; Feature extraction; Optical fiber networks; Optical filters; Optical imaging

FlowNet2 [14], the state-of-the-art convolutional neural network (CNN) for optical flow estimation, requires over 160M parameters to achieve accurate flow estimation. In this paper we present an alternative network that attains performance on par with FlowNet2 on the challenging Sintel final pass an...