AI Paper Atlas

Research map / Detection and segmentation / CNN features

Structured Binary Neural Networks for Image Recognition

Bohan Zhuang, Chunhua Shen, Mingkui Tan, Peng Chen, Lingqiao Liu, Ian D. Reid

IJCV 2022 · Detection and segmentation › CNN features · approach: Efficient network training and design · 26 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
add learned soft gates (ablation)GroupNet-A (ResNet-34) · ImageNet valTop-1 accuracy: 66.4 -> 68.7Improved
add learned soft gates (ablation)GroupNet-A (ResNet-18) · ImageNet valTop-1 accuracy: 64.1 -> 65.2Improved
add learned soft gates (ablation)GroupNet-A (ResNet-50) · ImageNet valTop-1 accuracy: 67.7 -> 69.8Improved
keep 1x1 downsampling skip connections at 8-bit (GroupNet-B vs A)GroupNet · ImageNet val (ResNet-18)Top-1 accuracy: 65.2 -> 67.3Improved
replace soft routing with dynamic hard gating (GroupNet-C vs B)GroupNet · ImageNet val (ResNet-18)Top-1 accuracy: 67.3 -> 68.2Improved
group decomposition instead of width multiplier 2 (ablation)GroupNet-A · ImageNet val (ResNet-18)Top-1 accuracy: 63.5 -> 65.2Improved
group-wise (block) decomposition instead of layer-wise (ablation)GroupNet-A · ImageNet val (ResNet-18)Top-1 accuracy: 57.8 -> 65.2Improved
one block per base instead of whole network per base (ablation)GroupNet · ImageNet val (ResNet-18)Top-1 accuracy: 59.8 -> 65.2Improved
increase number of bases from 1 to 5 (ablation)GroupNet-A · ImageNet val (ResNet-18)Top-1 accuracy: 58.1 -> 66.0Improved
use diverse dilation rates across binary bases (BPAC)GroupNet-C · PASCAL VOC 2012 val (ResNet-18 FCN-8s-C5)mIOU: 61.5 -> 66.2Improved
binarize ASPP activations to {0,1} instead of {-1,1} (ablation)binary DeepLabv3 · PASCAL VOC 2012 val (ResNet-18)mIOU: 51.6 -> 64.0Improved
binarize the ASPP module in addition to the backbone (ablation)binary DeepLabv3 · PASCAL VOC 2012 val (ResNet-34)mIOU: 71.9 -> 68.0Hurt

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