Test suite of 18 challenging pairs of images for testing registration algorithms
Our goal is an automated registration algorithm capable of aligning image pairs having some combination of low overlap, substantial orientation and scale differences, large illumination differences (e.g. day and night), substantial scene changes, and different modalities.
http://www.vision.cs.rpi.edu/gdbicp/dataset/

Generalized Dual Bootstrap-ICP executable
http://www.vision.cs.rpi.edu/gdbicp/exec/


Retinal Vessel Centerline Tracing executable
http://www.vision.cs.rpi.edu/vessels/


Test suite of 22 image pairs for testing multimodal registration and for keypoint analysis
http://www.vision.cs.rpi.edu/keypoints/


Dual-Bootstrap ICP

Dual-Bootstrap ICP ppt : containing animations and examples of the Dual-Bootstrap ICP registration algorithm in the retina application. [22Mb]

Change animation ppt : showing the use of precise registration in detecting longitudinal changes in a patient's retina [4.7Mb]


C++ implementation of the MUSE (Minimum Unbiased Scale Estimate) robust local surface patch extractor. Divides input range image into a series of non-overlapping windows and extracts a set of surfaces from each window. A reconstructed image, a surface description file (in IBM Visualization Data Explorer format), and a segmentation image are provided on output. (267KB gzipped tarfile) For more information, see publications on MUSE.
muse_v1.6.tar.gz

 

 

Home | People | Research | Publications | Contact | News/Events | Download | Links