Tensor Voting with Steerable Filters

Tool to do fast perceptual grouping on 2D images by tensor voting

Tensor voting is a framework for feature inference in images, based on perceptual principles. This tool implements a steerable filter formulation for 2D dense Tensor Voting. It serves as complimentary material to the following literature review on the framework:
"Perceptual grouping by tensor voting: a comparative survey of recent approaches". E Maggiori, HL Manterola, M del fresno. To be published in IET Computer Vision.

In particular, this tool implements a tensor voting reformulation which runs extremely fast compared to previous work. This formulation was presented in:

"An efficient method for tensor voting using steerable filters", Franken et. al. ECCV 2006.