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.

TransFlowNet: A Physics-Constrained Transformer Framework for Spatio-Temporal Super-Resolution of Flow Simulations.

TransFlow: Transformer as Flow Learner

Super-resolution reconstruction of turbulent flows with a transformer-based deep learning framework

A generative adversarial network based on an efficient transformer for high-fidelity flow field reconstruction

Optical Flow as Spatial-Temporal Attention Learners

PhySR: Physics-informed Deep Super-resolution for Spatiotemporal Data

A Swin-transformer-based model for efficient compression of turbulent flow data

Deep Learning Methods for Super-Resolution Reconstruction of Turbulent Flows

Deep learning method for super-resolution reconstruction of the spatio-temporal flow field

Physics-driven Learning of the Steady Navier-Stokes Equations using Deep Convolutional Neural Networks

Physics Constrained Flow Neural Network for Short-Timescale Predictions in Data Communications Networks

Physics-constrained and flow-field-message-informed graph neural network for solving unsteady compressible flows

LoLiSRFlow: Joint Single Image Low-light Enhancement and Super-resolution via Cross-scale Transformer-based Conditional Flow

High-Resolution Urban Flows Forecasting with Coarse-Grained Spatiotemporal Data

SDformerFlow: Spatiotemporal swin spikeformer for event-based optical flow estimation

Resistance-Time Co-Modulated PointNet for Temporal Super-Resolution Simulation of Blood Vessel Flows

RFRFlow: Recurrent feature refinement network for optical flow estimation

PiRD: Physics-informed Residual Diffusion for Flow Field Reconstruction

Unsupervised Denoising and Super-Resolution of Vascular Flow Data by Physics-informed Machine Learning

PSRFlow: Probabilistic Super Resolution with Flow-Based Models for Scientific Data