Premium
Fast ANN for High‐Quality Collaborative Filtering
Author(s) -
Tsai YunTa,
Steinberger Markus,
Pająk Dawid,
Pulli Kari
Publication year - 2016
Publication title -
computer graphics forum
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.578
H-Index - 120
eISSN - 1467-8659
pISSN - 0167-7055
DOI - 10.1111/cgf.12715
Subject(s) - computer science , collaborative filtering , filter (signal processing) , artificial intelligence , similarity (geometry) , pixel , matching (statistics) , image (mathematics) , quality (philosophy) , pattern recognition (psychology) , computer vision , algorithm , machine learning , recommender system , mathematics , philosophy , statistics , epistemology
Collaborative filtering collects similar patches, jointly filters them and scatters the output back to input patches; each pixel gets a contribution from each patch that overlaps with it, allowing signal reconstruction from highly corrupted data. Exploiting self‐similarity, however, requires finding matching image patches, which is an expensive operation. We propose a GPU‐friendly approximated‐nearest‐neighbour(ANN) algorithm that produces high‐quality results for any type of collaborative filter. We evaluate our ANN search against state‐of‐the‐art ANN algorithms in several application domains. Our method is orders of magnitudes faster, yet provides similar or higher quality results than the previous work.