activity
20202022
most citedMDFlow: Unsupervised Optical Flow Learning by Reliable Mutual Knowledge Distillation

36 citations · 57 across the 7 of their papers we have counts for

collaborators

7 papers

cs.CV20228 cited

Progressive Motion Context Refine Network for Efficient Video Frame Interpolation

Lingtong Kong, Jinfeng Liu, Jie Yang

Recently, flow-based frame interpolation methods have achieved great success by first modeling optical flow between target and input frames, and then building synthesis network for…

cs.CV202236 cited

MDFlow: Unsupervised Optical Flow Learning by Reliable Mutual Knowledge Distillation

Lingtong Kong, Jie Yang

Recent works have shown that optical flow can be learned by deep networks from unlabelled image pairs based on brightness constancy assumption and smoothness prior. Current approac…

cs.CV20226 cited

IFRNet: Intermediate Feature Refine Network for Efficient Frame Interpolation

Lingtong Kong, Boyuan Jiang, Donghao Luo +5

Prevailing video frame interpolation algorithms, that generate the intermediate frames from consecutive inputs, typically rely on complex model architectures with heavy parameters…

cs.CV20215 cited

FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation

Lingtong Kong, Chunhua Shen, Jie Yang

Dense optical flow estimation plays a key role in many robotic vision tasks. In the past few years, with the advent of deep learning, we have witnessed great progress in optical fl…

cs.CV2021

Unsupervised Motion Representation Enhanced Network for Action Recognition

Xiaohang Yang, Lingtong Kong, Jie Yang

Learning reliable motion representation between consecutive frames, such as optical flow, has proven to have great promotion to video understanding. However, the TV-L1 method, an e…

cs.CV2021

OAS-Net: Occlusion Aware Sampling Network for Accurate Optical Flow

Lingtong Kong, Xiaohang Yang, Jie Yang

Optical flow estimation is an essential step for many real-world computer vision tasks. Existing deep networks have achieved satisfactory results by mostly employing a pyramidal co…