… that supervised CNNs can learn to estimate optical flow well even when only guided using
noisy … Our proposed guided learning is completely unsupervised and achieves competitive or …

… pyramid learning framework of unsupervised optical flow estimation. … -guided upsampling
module that supports blur-free optical flow upsampling by using a selflearned interpolation flow …

… an unsupervised learning approach that fuses gyroscope into optical flow learning. Specifically,
… Second, we design a self-guided fusion module to fuse the background motion extracted …

… guided by the motion information. Taking the above requirements into consideration, we
propose the Optical Flow guided … from the orthogonal space of optical flow on the feature level […

… learning approach that fuses gyroscope into homography and optical flow learning. … the
gyro field and the image-based optical flow, we propose a self-guided fusion module (SGF). On …

… the problem of unsupervised learning of optical flow using a … methods on optical flow
estimation and learning from video. Our … Optical flow Optical flow is a classic problem in computer …

… In this paper, we make use of the domain knowledge of optical flow, and take reliable
predictions of non-occluded pixels as the self-supervision signal to guide our optical flow learning …

… predicting pixel-wise foreground object segmentation and optical flow in videos. • We
demonstrate that optical flow and video object segmentation tasks are complementary, and can …

… distillation approach to learning optical flow estimation from … annotations to guide a student
network to learn optical flow. … is data-driven, and learns optical flow for occluded pixels. This …

… the guided optical flow framework. Supposing the availability of a sparse yet accurate set
of optical flow … to guide them towards more accurate results. To this aim, we first extend the …