Related Experiment Video
Updated: Jun 5, 2026
07:15
Transient Optical Clearing Using Absorbing Molecules for Ex Vivo and In Vivo Imaging
Published on: July 11, 2025
Removal of partial occlusion from single images
Scott McCloskey1, Michael Langer, Kaleem Siddiqi
1School of Computer Science, McGill University, Montreal, QC, Canada. scott@cim.mcgill.ca
IEEE Transactions on Pattern Analysis and Machine Intelligence
|January 22, 2011
Summary
This study presents a novel method to remove foreground occlusions in images, enhancing background visibility. The technique effectively handles out-of-focus objects and depth discontinuities with minimal user input.
Area of Science:
- Computer Vision
- Image Processing
- Computational Photography
Background:
- Partial occlusions near depth discontinuities pose challenges in image analysis.
- Severely out-of-focus foreground objects complicate occlusion modeling.
- Existing methods often require significant user interaction.
Purpose of the Study:
- To develop a method for removing foreground occluder contributions in partially occluded image regions.
- To improve the visibility of background scenes obscured by out-of-focus foreground objects.
- To enable single-image occlusion removal with minimal user input.
Main Methods:
- Modeling partial occlusions using matting with alpha values derived from blur kernel convolution.
- Estimating the region of complete occlusion via curve evolution.
- Estimating alpha values for pixels in partly occluded regions.
- Removing foreground occluder intensity contributions.
Main Results:
- Successfully removed the image contribution of foreground occluders in regions of partial occlusion.
- Significantly improved the visibility of background scenes.
- Demonstrated effectiveness in single images with minimal user intervention.
Conclusions:
- The proposed method effectively handles large partial occlusions caused by out-of-focus foreground objects.
- It offers an automated approach to enhance background visibility in challenging occlusion scenarios.
- The technique shows promise for various image processing and computer vision applications.
