AI Paper Atlas

Research map / Detection and segmentation / CNN features

Attend and Rectify: a gated attention mechanism for fine-grained recovery

Pau Rodriguez, Josep M. Gonfaus, Guillem Cucurull, F. XavierRoca, Jordi Gonzalez

ECCV 2018 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 45 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
augment WRN-28 with attention (WARN, no dropout, AD2 AW4)WRN-28 · CIFAR-10 (from scratch)error rate (%): 4 -> 3.60Improved
augment WRN-28 with attention (WARN, AD3 AW4, dropout)WRN-28 · CIFAR-100 (from scratch)error rate (%): 18.85 -> 18.26Improved
augment WRN-40 with attention (WARN)WRN-40 · CIFAR-100 (from scratch)error rate (%): 18.3 -> 17.82Improved
use WARN attention instead of WRN-att2WRN-40 · CIFAR-100 (from scratch)error rate (%): 19.20 -> 17.82Improved
increase attention depth from 1 to number of residual blocks and width from 1 to 4 (ablation)WARN · CIFAR-100 (from scratch)error rate (%): 1.5% accuracyImproved
use gating instead of linear projection (ablation)WARN · CIFAR-100 (from scratch)accuracy: 4% betterImproved
use gating instead of averaging output vectors (ablation)WARN · CIFAR-100 (from scratch)accuracy: 3% betterImproved
add regularization of attention heads (ablation)WARN · CIFAR-100 (from scratch)accuracy: 0.6% improvementImproved
use sigmoid attention instead of softmax (ablation)WARN · CIFAR-100 (from scratch)accuracyNo clear effect
allow gradients from attention modules to the base network (ablation)WARN · CIFAR-10 (from scratch)accuracy: 0.2% drop when blockedImproved
add gates, softmax and regularization (best configuration) (ablation)CNN + attention · Cluttered Translated MNISTaccuracy (%): 97.8Improved
add attention to ImageNet-pretrained WRN-50-4WRN-50-4 · Stanford Dogs (fine-tuned)accuracy (%): 89.6 -> 92.9Improved

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