AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Crowd Counting With Deep Negative Correlation Learning

Zenglin Shi, Le Zhang, Yun Liu, Xiaofeng Cao, Yangdong Ye, Ming-Ming Cheng, Guoyan Zheng

CVPR 2018 · Architectures and efficiency › CNN design · approach: Filter design and trained-filter analysis · 312 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 negative-correlation regularization to the conventional ensemble (ablation)D-ConvNet-v1 · UCF CC 50MAE: 380.5 -> 288.4Improved
Add negative-correlation regularization to the conventional ensemble (ablation)D-ConvNet-v1 · UCF CC 50RMSE: 527.2 -> 404.7Improved
Add negative-correlation regularization to the conventional ensemble (ablation)D-ConvNet-v1 · ShanghaiTech Part AMAE: 91.6 -> 73.5Improved
Add negative-correlation regularization to the conventional ensemble (ablation)D-ConvNet-v1 · ShanghaiTech Part BMAE: 21.3 -> 18.7Improved
Add negative-correlation regularization to the conventional ensemble (ablation)D-ConvNet-v1 · WorldExpo'10MAE: 16.4 -> 9.1Improved
Use the customized shallower network instead of VGG16 (ablation)D-ConvNet · UCF CC 50MAE: 288.4 -> 354.1Hurt
Use the customized shallower network instead of VGG16 (ablation)D-ConvNet · ShanghaiTech Part AMAE: 73.5 -> 101.7Hurt
Use the customized shallower network instead of VGG16 (ablation)D-ConvNet · ShanghaiTech Part BMAE: 18.7 -> 25.7Hurt
Use the customized shallower network instead of VGG16 (ablation)D-ConvNet · WorldExpo'10 averageMAE: 9.1 -> 10.7Hurt
Set NCL correlation coefficient λ to a relatively smaller value in [10−3, 10−2] (ablation)D-ConvNet · Crowd counting datasetsMAENo clear effect
Increase ensemble size beyond 64 regressors (ablation)D-ConvNet · Crowd counting datasetsMAENo clear effect

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