Fast ANN for High‐Quality Collaborative Filtering |
| |
Authors: | Yun‐Ta Tsai Markus Steinberger Dawid Paj?k Kari Pulli |
| |
Affiliation: | 1. NVIDIA, Santa Clara, US;2. Google Inc., Mountain View, US;3. Graz University of Technology, Graz, Austria;4. Light, Palo Alto, US |
| |
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‐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. |
| |
Keywords: | approximated nearest neighborhood parallel computing non‐local means denoising I 4 3 [Image Processing and Computer Vision]: Enhancement Filtering |
|
|