AI Paper Atlas

Research map / 3D vision / Human pose estimation

3D-POP - An Automated Annotation Approach to Facilitate Markerless 2D-3D Tracking of Freely Moving Birds With Marker-Based Motion Capture

Hemal Naik, Alex Hoi Hang Chan, Junran Yang, Mathilde Delacoux, Iain D. Couzin, Fumihiro Kano, Máté Nagy

CVPR 2023 · 3D vision › Human pose estimation · approach: Physics and sensor-based motion capture · 33 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 DLC-detected keypoints from four views and triangulate to fill missing mo-cap trackingHybrid marker-based and markerless tracking · Single-pigeon sequence with 25% of mo-cap data removedavg. RMS error: 9.2 mm vs. 52.1 mmImproved
Add outlier-frame filtering at inference (ablation)DLC ResNet50 keypoint model · Beak 2D keypointRMSE: 10.1 -> 8.1Improved
Train on filtered data rather than the full dataset (ablation)DLC ResNet50 keypoint model · Beak 2D keypointRMSE: 8.1 -> 8.4Hurt
Use the 3D-POP annotation pipeline to estimate body keypointsBody keypoint annotations · 1000 randomly selected framesPCK05: 66%Mixed
Use the 3D-POP annotation pipeline to estimate body keypointsBody keypoint annotations · 1000 randomly selected framesPCK10: 94%Improved

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