On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,982 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “PatchMatch”

SCRAM: Spatially Coherent Randomized Attention Mapscs.LG

SCRAM accelerates image attention computation to O(n log(n)) from O(n^2).

problem Efficiently computing attention maps for large images.

method SCRAM uses PatchMatch to quickly find key patches and then approximates non-local mean operations.

result SCRAM achieves O(n log(n)) time complexity for image attention computation.

attention maps image processing PatchMatch non-local mean Transformer

2019-05-24abs ↗pdf ↗

Fast Feature Extraction with CNNs with Pooling Layerscs.CV

In recent years, many publications showed that convolutional neural network based features can have a superior performance to engineered features. However, not much effort was taken so far to extract local features efficiently for a whole image. In this paper, we present an approach to compute patch-based local feature…

2018-05-08abs ↗pdf ↗