AI Paper Atlas

Research map / Detection and segmentation / CNN features

Locate, Size and Count: Accurately Resolving People in Dense Crowds via Detection

Deepak Babu Sam, Skand Vishwanath Peri, Mukuntha Narayanan Sundararaman, Amogh Kamath, R. Venkatesh Babu

TPAMI 2020 · Detection and segmentation › CNN features · approach: Pedestrian and crowd detection · 246 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
use 4 scale branches instead of 1 (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 155.2 -> 66.4Improved
add a fifth scale branch (nS=5 vs 4) (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 66.4 -> 69.3Hurt
use 3 boxes per scale instead of 1 (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 104.7 -> 66.4Improved
use 4 boxes per scale instead of 3 (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 66.4 -> 74.3Hurt
add top-down feature modulation (multiple connections) (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 94.5 -> 66.4Improved
use connections from all lower scales instead of sequential only (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 73.4 -> 66.4Improved
concatenate top-down features instead of multiplicative gating (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 67.6 -> 66.4No clear effect
add GWTA training (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 79.2 -> 66.4Improved
add class weighting of the loss (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 360.1 -> 66.4Improved
replicate VGG blocks in the feature extractor (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 79.3 -> 66.4Improved
classify discretized box sizes per pixel instead of regressing sizes (ablation)LSC-CNN · ShanghaiTech Part_AMAE: 77.9 -> 66.4Improved
detection output with LSC-CNN instead of thresholded density regressionLSC-CNN vs CSR-A-thr · ShanghaiTech Part_AMLE (pixels): 16.8 -> 9.6Improved

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