AI Paper Atlas

Research map / Detection and segmentation / CNN features

Adaptive Sparse Convolutional Networks With Global Context Enhancement for Faster Object Detection on Drone Images

Bowei Du, Yecheng Huang, Jiaxin Chen, Di Huang

CVPR 2023 · Detection and segmentation › CNN features · approach: Context for hard detection conditions · 250 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
replace the GFL V1 detection head with CEASCGFL V1, ResNet18 · VisDroneGFLOPs: 524.95 -> 150.18Improved
replace the GFL V1 detection head with CEASCGFL V1, ResNet18 · VisDronemAP: 28.4 -> 28.7Improved
replace the detection head with CEASCGFL V1 · UAVDTmAP: 16.9 -> 17.1Improved
add CESC to the dense baseline (ablation)GFL V1 · VisDroneGFLOPs: 524.95 -> 158.23Improved
add adaptive multi-layer masking to CESC (ablation)GFL V1 with CESC · VisDronemAP: 28.6 -> 28.7Improved
add CE-GN to sparse convolution with a residual structure (ablation)GFL V1 with sparse convolution · VisDronemAP: 26.1 -> 28.5Improved
add normalization loss to sparse convolution, residual, and CE-GN (ablation)GFL V1 with CESC · VisDronemAP: 28.5 -> 28.7Improved
replace CE-GN with GroupNorm (ablation)GFL V1 with sparse convolution · VisDronemAP: 28.7 -> 28.0Hurt
replace CE-GN with BatchNorm (ablation)GFL V1 with sparse convolution · VisDronemAP: 28.7 -> 26.1Hurt
replace CE-GN with InstanceNorm (ablation)GFL V1 with sparse convolution · VisDronemAP: 28.7 -> 27.9Hurt
replace point-wise global-context encoding with 3×3 convolution (ablation)GFL V1 with CESC · VisDronemAP: 28.7 -> 28.5Hurt
replace point-wise global-context encoding with GhostModule (ablation)GFL V1 with CESC · VisDronemAP: 28.7 -> 28.3Hurt

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