AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Unrolled Memory Inner-Products: An Abstract GPU Operator for Efficient Vision-Related Computations

Yu-Sheng Lin, Wei-Chao Chen, Shao-Yi Chien

ICCV 2017 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use UMI instead of OpenCV separable filter, short kernelUMI GPU kernel · Separable filter k=3speedup ratio: 0.35Hurt
use UMI instead of OpenCV separable filter, long kernelUMI GPU kernel · Separable filter k=30speedup ratio: 1.42Improved
use UMI instead of Parboil motion estimationUMI GPU kernel · Motion estimationspeedup ratio: 6.51Improved
use UMI instead of Caffe for forward propagation, 3 + 1sUMI GPU kernel · Forward propagationspeedup ratio: 19.9Improved
use UMI instead of Caffe for forward propagation, 9 + 1sUMI GPU kernel · Forward propagationspeedup ratio: 26.4Improved
use UMI instead of Caffe for forward propagation, 3 + 2sUMI GPU kernel · Forward propagationspeedup ratio: 1.80Improved
use cuDNN instead of UMI for forward propagation, 3 + 1scuDNN · Forward propagationspeedup ratio: 19.9 -> 100Improved
express motion estimation with UMI instead of naive CPU codeUMI · Motion estimationoperator tokens: 45 -> 11Improved
express forward propagation with UMI instead of naive CPU codeUMI · Forward propagationidentifier tokens: 110 -> 53Improved
express bilateral filter with UMI instead of naive CPU codeUMI · Bilateral filterliteral tokens: 7 -> 61Hurt
express GEMM with UMI instead of naive CPU codeUMI · GEMMidentifier tokens: 34 -> 34No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy