no code implementations • 4 Aug 2019 • Xiaochuan Yin, Chengju Liu
In contrast to the previous methods, our proposed method calculates the camera motion with a direct method rather than regressing the ego-motion from the pose network.
no code implementations • 30 May 2018 • Henglai Wei, Xiaochuan Yin, Penghong Lin
Most of the existing methods for video prediction attempt to generate future frames in simple and fixed scenes.
1 code implementation • 20 Feb 2018 • Xiaochuan Yin, Henglai Wei, Penghong Lin, Xiangwei Wang, Qijun Chen
Novel view synthesis aims to synthesize new images from different viewpoints of given images.
no code implementations • ICCV 2017 • Xiaochuan Yin, Xiangwei Wang, Xiaoguo Du, Qijun Chen
Normally, road plane and camera height are specified as reference to recover the scale.