Abstract:Optical flow represents vector motion of targets, and its estimation is a hot research in computer vision. The traditional prior-based methods build mathematical models using different regularizers on object characteristics. These methods are straightforward and easily understood, but they suffer from a high computational burden. Deep learning based approaches greatly improve their efficiency, but their filter sizes are fixed and receptive fields are large. To solve these problems, this paper proposes a novel recurrent feature refinement network based on short-term dense connection modules for optical flow estimation (RFRFlow), including the following key parts: firstly, we utilize short-term dense concatenate modules (STDC) as a backbone part for constructing a feature extracting network which can preserve rich details and use global averaging pooling to extract global context information with a large receptive field. Secondly, a 4D correlation volume is constructed using context features of image pairs to extract multi-scale 4D correlation information. Thirdly, a motion feature encoder is reported to transfer the multi-scale 4D correlation information into 2D motion features. Finally, 2D context features and 2D motion features are input into a global motion aggregation (GMA) decoder for pursuing the final optical flow. In the experiments, we firstly present and analyze the results of ablation studies on the selection of batch size and image size. Then, both quantitative and qualitative results on Sintel and KITTI datasets are compared, showing that the proposed RFRFlow is effective and even performs better than the state-of-the-art optical flow estimation methods on performance improvement and objects’ shape preservation.

ReFlowNet: Revisiting Coarse-to-fine Learning of Optical Flow

A Dense Optical Flow Registration Algorithm Based on Deep Learning

Unsupervised Learning of Scene Flow Estimation Fusing with Local Rigidity.

RFRFlow: Recurrent feature refinement network for optical flow estimation

Optical Flow Estimation Using a Non-Local Convolutional Network.

FPCR-Net: Feature pyramidal correlation and residual reconstruction for optical flow estimation

MRDFlow: Unsupervised Optical Flow Estimation Network With Multi-Scale Recurrent Decoder

Self-Attention-Based Multiscale Feature Learning Optical Flow with Occlusion Feature Map Prediction

OFR-Net: Optical Flow Refinement with a Pyramid Dense Residual Network.

LCIF-Net: Local criss-cross attention based optical flow method using multi-scale image features and feature pyramid

Recurrent Partial Kernel Network for Efficient Optical Flow Estimation

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

FlowNet: Learning Optical Flow with Convolutional Networks

Rethinking Optical Flow from Geometric Matching Consistent Perspective

FastFlowNet: A Lightweight Network for Fast Optical Flow Estimation

Optical Flow Super-Resolution Based on Image Guidence Using Convolutional Neural Network

Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimation

Masked Label Learning For Optical Flow Regression

MemFlow: Optical Flow Estimation and Prediction with Memory