AI Paper Atlas

Research map / Video / Video segmentation and tracking

Counting People by Estimating People Flows

Weizhe Liu, Mathieu Salzmann, Pascal Fua

TPAMI 2021 · Video › Video segmentation and tracking · approach: Learned optical flow estimation · 40 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
regress people flows between locations instead of densities (ablation)CAN backbone · CrowdFlowMAE: 124.3 -> 113.3Improved
replace flow regression with direct density regression from two images (ablation)Ours · CrowdFlowMAE: 97.8 -> 125.7Hurt
average densities over three images instead of using flows (ablation)CAN · CrowdFlowMAE: 124.3 -> 128.9Hurt
add weaker conservation loss on densities (ablation)CAN · CrowdFlowMAE: 124.3 -> 121.2Improved
add forward-backward cycle consistency constraints (ablation)Flow regressor · CrowdFlowMAE: 113.3 -> 97.8Improved
add optical flow regularization from PWC-Net (ablation)Ours (combined loss) · CrowdFlowMAE: 97.8 -> 96.3Improved
use ground-truth instead of estimated optical flowOurs · CrowdFlowMAE: 96.3 -> 90.9Improved
feed original images to the optical flow regressor in addition to density maps (ablation)Ours (PWC-Net optical flow) · CrowdFlowMAE: 96.3 -> 97.5Hurt
add spatial loss term to the fully supervised loss (ablation)Ours (combined loss) · CrowdFlowMAE: 97.8 -> 97.7No clear effect
average forward and backward flows rather than using one (ablation)Ours · UCSDMAE: 0.87 -> 0.86Improved
work in the ground plane instead of the image planeOurs (combined loss) · VeniceMAE: 15.0 -> 12.3Improved
annotate every 2nd frame instead of every frame (ablation)Ours (combined loss) · CrowdFlowMAE: 97.8 -> 98.1No 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