GenOpticalFlow: A Generative Approach to Unsupervised Optical Flow Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914414688468992 |
|---|---|
| author | Luo, Yixuan Qiao, Feng Xiong, Zhexiao Li, Yanjing Jacobs, Nathan |
| author_facet | Luo, Yixuan Qiao, Feng Xiong, Zhexiao Li, Yanjing Jacobs, Nathan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_22270 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GenOpticalFlow: A Generative Approach to Unsupervised Optical Flow Learning Luo, Yixuan Qiao, Feng Xiong, Zhexiao Li, Yanjing Jacobs, Nathan Computer Vision and Pattern Recognition 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. |
| title | GenOpticalFlow: A Generative Approach to Unsupervised Optical Flow Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.22270 |