Skip to main content

Batch image alignment using the technique described in "Robust Alignment by Sparse and Low-rank Decomposition for Linearly Correlated Images"

Project description

Align linearly correlated images, possibly having gross corruption or occlusions.

Detailed description and installation instructions, along with example code and data, are here: https://github.com/welch/rasl

rasl is a python implementation of the batch image alignment technique described in:

  1. Peng, A. Ganesh, J. Wright, W. Xu, Y. Ma, “Robust Alignment by Sparse and Low-rank Decomposition for Linearly Correlated Images”, IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) 2011

The paper describes a technique for aligning images of objects varying in illumination and projection, possibly with occlusions (such as facial images at varying angles, some including eyeglasses or hair). RASL seeks transformations or deformations that will best superimpose a batch of images, with pixel accuracy where possible. It solves this problem by decomposing the image matrix into a dense low-rank component (analogous to “eigenfaces” in face-recognition literature) combined with a sparse error matrix representing any occlusions. The decomposition is accomplished with a robust form of PCA via Principal Components Pursuit.

Dependencies

numpy, scipy, scikit-image

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Filename, size & hash SHA256 hash help File type Python version Upload date
rasl-0.1.0-py2.py3-none-any.whl (18.5 kB) Copy SHA256 hash SHA256 Wheel 2.7 Apr 25, 2016
rasl-0.1.0.tar.gz (13.6 kB) Copy SHA256 hash SHA256 Source None Apr 25, 2016

Supported by

Elastic Elastic Search Pingdom Pingdom Monitoring Google Google BigQuery Sentry Sentry Error logging AWS AWS Cloud computing DataDog DataDog Monitoring Fastly Fastly CDN DigiCert DigiCert EV certificate StatusPage StatusPage Status page