AI Paper Atlas

Research map / Architectures and efficiency / CNN design

InterActive: Inter-Layer Activeness Propagation

Lingxi Xie, Liang Zheng, Jingdong Wang, Alan L. Yuille, Qi Tian

CVPR 2016 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 42 citations (OpenAlex)

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
propagate final-layer activeness to low- and mid-level featuresVGGNet features · pool-1 through pool-4, six classification datasetsaccuracyImproved
use final-layer rather than next-layer activeness supervision (ablation)InterActive · Caltech256, Indoor-67, SUN-397, Pet-37, Flower-102, Bird-200accuracy: 84.64 -> 85.06; 78.23 -> 78.65; 62.50 -> 62.97; 93.22 -> 93.45; 96.26 -> 96.40; 74.61 -> 75.62Improved
use p=2 rather than p=1 in final-layer activeness (ablation)InterActive · Caltech256, Indoor-67, SUN-397, Pet-37, Flower-102, Bird-200accuracy: 84.94 -> 85.06; 78.40 -> 78.65; 62.69 -> 62.97; 93.40 -> 93.45; 96.35 -> 96.40; 75.47 -> 75.62Improved
replace InterActive weighting with original average-pooled features (ablation)InterActive concatenated features · Caltech256, Indoor-67, SUN-397, Pet-37, Flower-102, Bird-200accuracy: 84.02 -> 85.06; 78.02 -> 78.65; 62.30 -> 62.97; 93.02 -> 93.45; 95.70 -> 96.40; 73.35 -> 75.62Improved
replace InterActive weighting with original max-pooled features (ablation)InterActive concatenated features · Caltech256, Indoor-67, SUN-397, Pet-37, Flower-102, Bird-200accuracy: 84.38 -> 85.06; 77.32 -> 78.65; 61.87 -> 62.97; 93.20 -> 93.45; 95.98 -> 96.40; 74.76 -> 75.62Improved
apply InterActive to high-level featuresVGGNet pool-5 and fc-6 features · scene recognitionaccuracyHurt
apply InterActive to high-level featuresVGGNet pool-5 and fc-6 features · object recognitionaccuracyMixed
use average pooling instead of max pooling (ablation)original deep features · scene recognitionaccuracyImproved
use max pooling instead of average pooling (ablation)original deep features · Bird-200accuracyImproved
preserve image aspect ratio instead of warping input images (ablation)VGGNet deep-feature extraction · Caltech256, SUN-397, Flower-102, fc-6accuracy: 80.41% -> 83.51%; 53.06% -> 61.30%; 84.89% -> 93.54%Improved
apply InterActive response update to fc-7 and recompute fc-8 scoresVGG16, VGG19, and combined VGG models · ILSVRC2012 validationerror rate: 7.1% -> 6.8%; 7.0% -> 6.7%; 6.7% -> 6.5%Improved
add object bounding boxes (ablation)original features and InterActive features · Bird-200accuracy: 76.95% -> 77.53%Improved

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