Abstract:The accuracy of learning-based optical flow estimation models heavily relies on the realism of the training datasets. Current approaches for generating such datasets either employ synthetic data or generate images with limited realism. However, the domain gap of these data with real-world scenes constrains the generalization of the trained model to real-world applications. To address this issue, we investigate generating realistic optical flow datasets from real-world images. Firstly, to generate highly realistic new images, we construct a layered depth representation, known as multiplane images (MPI), from single-view images. This allows us to generate novel view images that are highly realistic. To generate optical flow maps that correspond accurately to the new image, we calculate the optical flows of each plane using the camera matrix and plane depths. We then project these layered optical flows into the output optical flow map with volume rendering. Secondly, to ensure the realism of motion, we present an independent object motion module that can separate the camera and dynamic object motion in MPI. This module addresses the deficiency in MPI-based single-view methods, where optical flow is generated only by camera motion and does not account for any object movement. We additionally devise a depth-aware inpainting module to merge new images with dynamic objects and address unnatural motion occlusions. We show the superior performance of our method through extensive experiments on real-world datasets. Moreover, our approach achieves state-of-the-art performance in both unsupervised and supervised training of learning-based models. The code will be made publicly available at: \url{<a class="link-external link-https" href="https://github.com/Sharpiless/MPI-Flow" rel="external noopener nofollow">this https URL</a>}.

On Benchmarking Optical Flow

A Database and Evaluation Methodology for Optical Flow

Optical Flow Dataset and Benchmark for Visual Crowd Analysis

Dynamic Illumination Optical Flow Computing for Sensing Multiple Mobile Robots from a Drone.

A Quantitative Analysis of Current Practices in Optical Flow Estimation and the Principles Behind Them

Optical flow modeling and computation: A survey

Technical Report on Visual Quality Assessment for Frame Interpolation

Nonrigid Optical Flow Ground Truth for Real-World Scenes with Time-Varying Shading Effects

Optical Flow in Mostly Rigid Scenes

The devil in the details: simple and effective optical flow synthetic data generation

Robust Optical Flow Computation: A Higher-Order Differential Approach

Subjective Annotation for a Frame Interpolation Benchmark using Artefact Amplification

Spring: A High-Resolution High-Detail Dataset and Benchmark for Scene Flow, Optical Flow and Stereo

I Can't Believe It's Not Scene Flow!

OmniFlow: Human Omnidirectional Optical Flow

RealFlow: EM-Based Realistic Optical Flow Dataset Generation from Videos

HMAFlow: Learning More Accurate Optical Flow via Hierarchical Motion Field Alignment

MPI-Flow: Learning Realistic Optical Flow with Multiplane Images

BlinkVision: A Benchmark for Optical Flow, Scene Flow and Point Tracking Estimation using RGB Frames and Events

Optical Flow Computation Based On Bayesian Selecting Scheme

The Statistics of Optical Flow