Fast ANN for High-Quality Collaborative Filtering

Tsai, Yun-Ta; Steinberger, Markus; Pajak, Dawid; Pulli, Kari

Fast ANN for High-Quality Collaborative Filtering

Files

061-070.pdf (35.2 MB)

Date

2014

Authors

Tsai, Yun-Ta
Steinberger, Markus
Pajak, Dawid
Pulli, Kari

Publisher

The Eurographics Association

Abstract

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-neighbor 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.

        @inproceedings{:10.2312/hpg.20141094
https::/diglib.eg.org:443/handle/10.2312/hpg.20141094
,
booktitle = {Eurographics/ ACM SIGGRAPH Symposium on High Performance Graphics
},
editor = {Ingo Wald and Jonathan Ragan-Kelley
},
title = {{Fast ANN for High-Quality Collaborative Filtering
}},
author = {Tsai, Yun-Ta and 
Steinberger, Markus and 
Pajak, Dawid and 
Pulli, Kari
},
year = {2014
},
publisher = {The Eurographics Association
},
ISSN = {2079-8679
},
ISBN = {978-3-905674-60-6
},
DOI = {/10.2312/hpg.20141094
https://diglib.eg.org:443/handle/10.2312/hpg.20141094
}
}