# Learning to See Through Obstructions

Source: [Learning to See Through Obstructions](<https://www.sota2.com/research/paper/68fa67561e209131cea15a38>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:56:42.418Z

Authors: Yu-Lun Liu, Wei-Sheng Lai, Ming-Hsuan Yang, Yung-Yu Chuang, Jia-Bin Huang

Published: 2020-04-02T17:59:12.000Z

## Abstract

We present a learning-based approach for removing unwanted obstructions, such as window reflections, fence occlusions or raindrops, from a short sequence of images captured by a moving camera. Our method leverages the motion differences between the background and the obstructing elements to recover both layers. Specifically, we alternate between estimating dense optical flow fields of the two layers and reconstructing each layer from the flow-warped images via a deep convolutional neural network. The learning-based layer reconstruction allows us to accommodate potential errors in the flow estimation and brittle assumptions such as brightness consistency. We show that training on synthetically generated data transfers well to real images. Our results on numerous challenging scenarios of reflection and fence removal demonstrate the effectiveness of the proposed method.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Video waterdrop removal on Real driving scenes](<https://www.sota2.com/research/sota/video-waterdrop-removal-on-real-driving-scenes.md>): Video Waterdrop Removal; Real driving scenes; User Preference: 0.56.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2004.01180>)

[PDF](<https://arxiv.org/pdf/2004.01180.pdf>)