# Optical Flow in Dense Foggy Scenes using Semi-Supervised Learning

arXiv: https://arxiv.org/abs/2004.01905 · HTML: https://mlsanity.com/paper/2004.01905/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Apr 2020.

## Built on by (most cited first, 20 of 20)

- [FlowFormer: A Transformer Architecture for Optical Flow](https://mlsanity.com/paper/2203.16194/) (2022)
- [VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimation](https://mlsanity.com/paper/2303.08340/) (2023)
- [Unsupervised Night Image Enhancement: When Layer Decomposition Meets Light-Effects Suppression](https://mlsanity.com/paper/2207.10564/) (2022)
- [DC-ShadowNet: Single-Image Hard and Soft Shadow Removal Using Unsupervised Domain-Classifier Guided Network](https://mlsanity.com/paper/2207.10434/) (2022)
- [Imposing Consistency for Optical Flow Estimation](https://mlsanity.com/paper/2204.07262/) (2022)
- [GyroFlow: Gyroscope-Guided Unsupervised Optical Flow Learning](https://mlsanity.com/paper/2103.13725/) (2021)
- [Distracting Downpour: Adversarial Weather Attacks for Motion Estimation](https://mlsanity.com/paper/2305.06716/) (2023)
- [Optical Flow Training under Limited Label Budget via Active Learning](https://mlsanity.com/paper/2203.05053/) (2022)
- [GyroFlow+: Gyroscope-Guided Unsupervised Deep Homography and Optical Flow Learning](https://mlsanity.com/paper/2301.10018/) (2023)
- [Semi-Supervised Learning of Optical Flow by Flow Supervisor](https://mlsanity.com/paper/2207.10314/) (2022)
- [Optical Flow for Autonomous Driving: Applications, Challenges and Improvements](https://mlsanity.com/paper/2301.04422/) (2023)
- [Unsupervised Cumulative Domain Adaptation for Foggy Scene Optical Flow](https://mlsanity.com/paper/2303.07564/) (2023)
- [LIFE: Lighting Invariant Flow Estimation](https://mlsanity.com/paper/2104.03097/) (2021)
- [RVSL: Robust Vehicle Similarity Learning in Real Hazy Scenes Based on Semi-supervised Learning](https://mlsanity.com/paper/2209.08630/) (2022)
- [Adverse Weather Optical Flow: Cumulative Homogeneous-Heterogeneous Adaptation](https://mlsanity.com/paper/2409.17001/) (2024)
- [Unsupervised Hierarchical Domain Adaptation for Adverse Weather Optical Flow](https://mlsanity.com/paper/2303.13761/) (2023)
- [FlowFormer: A Transformer Architecture and Its Masked Cost Volume Autoencoding for Optical Flow](https://mlsanity.com/paper/2306.05442/) (2023)
- [Complementing Brightness Constancy with Deep Networks for Optical Flow Prediction](https://mlsanity.com/paper/2207.03790/) (2022)
- [Rethink Predicting the Optical Flow with the Kinetics Perspective](https://mlsanity.com/paper/2405.12512/) (2024)
- [GenOpticalFlow: A Generative Approach to Unsupervised Optical Flow Learning](https://mlsanity.com/paper/2603.22270/) (2026)

## Abstract (arXiv)

In dense foggy scenes, existing optical flow methods are erroneous. This is due to the degradation caused by dense fog particles that break the optical flow basic assumptions such as brightness and gradient constancy. To address the problem, we introduce a semi-supervised deep learning technique that employs real fog images without optical flow ground-truths in the training process. Our network integrates the domain transformation and optical flow networks in one framework. Initially, given a pair of synthetic fog images, its corresponding clean images and optical flow ground-truths, in one training batch we train our network in a supervised manner. Subsequently, given a pair of real fog images and a pair of clean images that are not corresponding to each other (unpaired), in the next training batch, we train our network in an unsupervised manner. We then alternate the training of synthetic and real data iteratively. We use real data without ground-truths, since to have ground-truths in such conditions is intractable, and also to avoid the overfitting problem of synthetic data training, where the knowledge learned on synthetic data cannot be generalized to real data testing. Together with the network architecture design, we propose a new training strategy that combines supervised synthetic-data training and unsupervised real-data training. Experimental results show that our method is effective and outperforms the state-of-the-art methods in estimating optical flow in dense foggy scenes.