AI Paper Atlas

Research map / Detection and segmentation / CNN features

Oriented Response Networks

Yanzhao Zhou, Qixiang Ye, Qiang Qiu, Jianbin Jiao

CVPR 2017 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 271 citations (OpenAlex)

Check 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
Replace standard convolutions with ORConv using ARFs and ORAlignCNN · MNIST-originalerror rate (%): 0.73 -> 0.57Improved
Replace standard convolutions with ARFs and ORPoolingCNN · MNIST-roterror rate (%): 2.82 -> 1.37Improved
Use ARFs with ORAlignCNN · MNIST-rot+error rate (%): 2.19 -> 1.12Improved
Train with ARFs and ORAlign instead of a conventional CNNCNN · MNIST-original to MNIST-roterror rate (%): 56.28 -> 16.21Improved
Add ORAlign to the OR-STN localization network (ablation)OR-STN · MNIST-half-rotclassification error (%): 2.43 -> 3.61Hurt
Upgrade the STN localization network from CNN to ORN without alignmentSTN · MNIST-half-rotorientation standard deviation: 0.745 -> 0.397Improved
Upgrade the STN localization network from CNN to ORN without alignmentSTN · MNIST-half-rotclassification error (%): 3.38 -> 2.43Improved
Replace the CNN with ORN-8(ORAlign)CNN · MNIST-rot-12kerror rate (%): 4.34 -> 2.25Improved
Replace the CNN branches in TIPooling with ORN-8TIPooling · MNIST-rot-12kerror rate (%): 1.93 -> 1.54Improved
Replace standard convolutions with ORConvVGG · CIFAR10classification error (%): 6.32 -> 5.47Improved
Replace standard convolutions with ORConvVGG · CIFAR100classification error (%): 28.49 -> 27.03Improved
Replace standard convolutions with ORConvResNet · CIFAR10classification error (%): 6.43 -> 5.31Improved

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