AI Paper Atlas

Research map / Detection and segmentation / CNN features

CTNet: Context-Based Tandem Network for Semantic Segmentation

Zechao Li, Yanpeng Sun, Liyan Zhang, Jinhui Tang

TPAMI 2021 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 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 SCM after CCM (tandem) instead of CCM only (ablation)CTNet (ResNet-101) · PASCAL-ContextmIoU (MS): 55.6 -> 56.2Improved
add CCM before SCM instead of SCM only (ablation)CTNet (ResNet-101) · PASCAL-ContextmIoU (MS): 54.5 -> 56.2Improved
add SCM to CCM (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU (MS): 52.6 -> 53.3Improved
add CCM to SCM (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU (MS): 52.3 -> 53.3Improved
connect CCM and SCM in tandem rather than in parallel (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU (MS): 52.6 -> 53.3Improved
connect CCM and SCM in tandem rather than in parallel (ablation)CTNet (ResNet-101) · PASCAL-ContextmIoU (MS): 55.4 -> 56.2Improved
use multi-scale MCE block instead of single-scale ECA block (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU: 52.2 -> 53.3Improved
use local channel context (ECA) instead of SE global context (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU: 51.8 -> 52.2Improved
fuse five kernel scales instead of four in MCE (ablation)CTNet MCE · PASCAL-ContextmIoU: 53.3 -> 52.4Hurt
use global average pooling instead of max pooling in CCM (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU: 52.3 -> 53.3Improved
use global average pooling instead of mix pooling in CCM (ablation)CTNet (ResNet-50) · PASCAL-ContextmIoU: 52.5 -> 53.3Improved
add CP-Loss (beta=0.1) to the loss (ablation)CTNet · PASCAL-ContextmIoUImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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