AI Paper Atlas

Research map / Detection and segmentation / CNN features

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

Liang-Chieh Chen, George Papandreou, Iasonas Kokkinos, Kevin P. Murphy, Alan Yuille

TPAMI 2017 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 23,273 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 dense CRF post-processing (ablation)DeepLab VGG-16 variants · PASCAL VOC 2012 valmIOU: 64.38 -> 67.64Improved
use LargeFOV (3x3 kernel, r=12) instead of 7x7 kernel (ablation)DeepLab VGG-16 · PASCAL VOC 2012 valmIOU after CRF: 67.64 -> 67.64No clear effect
use poly instead of step learning rate policy (ablation)DeepLab-LargeFOV · PASCAL VOC 2012 valmIOU: 62.25 -> 63.42Improved
train longer (6K -> 10K iterations) with poly (ablation)DeepLab-LargeFOV · PASCAL VOC 2012 valmIOU: 63.42 -> 64.90Improved
reduce batch size from 30 to 10 (ablation)DeepLab-LargeFOV · PASCAL VOC 2012 valmIOU: 64.90 -> 64.71No clear effect
add ASPP-S (rates 2,4,8,12) over LargeFOV (ablation)DeepLab VGG-16 · PASCAL VOC 2012 valmIOU: 65.76 -> 66.98 before CRF; 69.84 -> 69.73 after CRFMixed
add ASPP-L (rates 6,12,18,24) over LargeFOV (ablation)DeepLab VGG-16 · PASCAL VOC 2012 valmIOU: 65.76 -> 68.96 before CRF; 69.84 -> 71.57 after CRFImproved
replace VGG-16 with ResNet-101 (ablation)DeepLab · PASCAL VOC 2012 valmIOU before CRF: 65.76 -> 68.72Improved
add multi-scale inputs with max fusion (ablation)DeepLab ResNet-101 · PASCAL VOC 2012 valmIOU: 68.72 -> 71.27Improved
add MS-COCO pretraining (ablation)DeepLab ResNet-101 · PASCAL VOC 2012 valmIOU: 71.27 -> 73.28Improved
add random-scale data augmentation (ablation)DeepLab ResNet-101 · PASCAL VOC 2012 valmIOU: 73.28 -> 74.87Improved
add LargeFOV layer (ablation)DeepLab ResNet-101 · PASCAL VOC 2012 valmIOU: 74.87 -> 75.54Improved

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