GenOpticalFlow: A Generative Approach to Unsupervised Optical Flow Learning

TL;DR

GenOpticalFlow introduces a generative framework that synthesizes high-quality, aligned frame-flow data for unsupervised optical flow learning, reducing reliance on annotations and improving accuracy in real-world scenarios.

Contribution

The paper presents a novel data synthesis method using depth estimation and frame generation, along with an inconsistent pixel filtering strategy, to enhance unsupervised optical flow training.

Findings

Achieves competitive results on KITTI and Sintel datasets.

Outperforms existing unsupervised methods in accuracy.

Provides a scalable, annotation-free training pipeline.

Abstract

Optical flow estimation is a fundamental problem in computer vision, yet the reliance on expensive ground-truth annotations limits the scalability of supervised approaches. Although unsupervised and semi-supervised methods alleviate this issue, they often suffer from unreliable supervision signals based on brightness constancy and smoothness assumptions, leading to inaccurate motion estimation in complex real-world scenarios. To overcome these limitations, we introduce \textbf{\modelname}, a novel framework that synthesizes large-scale, perfectly aligned frame--flow data pairs for supervised optical flow training without human annotations. Specifically, our method leverages a pre-trained depth estimation network to generate pseudo optical flows, which serve as conditioning inputs for a next-frame generation model trained to produce high-fidelity, pixel-aligned subsequent frames. This process enables the creation of abundant, high-quality synthetic data with precise motion correspondence. Furthermore, we propose an \textit{inconsistent pixel filtering} strategy that identifies and removes unreliable pixels in generated frames, effectively enhancing fine-tuning performance on real-world datasets. Extensive experiments on KITTI2012, KITTI2015, and Sintel demonstrate that \textbf{\modelname} achieves competitive or superior results compared to existing unsupervised and semi-supervised approaches, highlighting its potential as a scalable and annotation-free solution for optical flow learning. We will release our code upon acceptance.