AI Paper Atlas

Research map / 3D vision / Human pose estimation

3D-MuPPET: 3D Multi-Pigeon Pose Estimation and Tracking

Urs Waldmann, Alex Hoi Hang Chan, Hemal Naik, Máté Nagy, Iain D. Couzin, Oliver Deußen, Bastian Goldlüecke, Fumihiro Kano

IJCV 2024 · 3D vision › Human pose estimation · approach: Temporal video pose estimation · 34 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 ViTPose* instead of KeypointRCNN as 2D pose module3D-MuPPET · 3D-POP 2DPCK05 (%): 82.4 -> 91.1Improved
use ViTPose* instead of KeypointRCNN as 2D pose module3D-MuPPET · 3D-POP 2Dmedian error (px): 5.7 -> 4.4Improved
use ViTPose* instead of KeypointRCNN as 2D pose module3D-MuPPET · 3D-POP 2Dmean speed (fps): 7.5 -> 2.1Hurt
use 3D-ViTPose* instead of 3D-KP-RCNN3D-MuPPET · 3D-POP 3DPCK05 (%): 53.2 -> 71.0Improved
use 3D-ViTPose* instead of 3D-KP-RCNN3D-MuPPET · 3D-POP 3Dmean speed (fps): 1.76 -> 0.51Hurt
use DLC* instead of KeypointRCNN3D-MuPPET · 3D-POP 3DPCK10 (%): 85.4 -> 90.9Improved
compare 3D-MuPPET (3D-ViTPose*) against LToHP, which uses 3D ground truth3D-ViTPose* · 3D-POP 3DRMSE (mm): 24.0 -> 14.8Hurt
compare 3D-MuPPET (3D-ViTPose*) against LToHP3D-ViTPose* · 3D-POP 3Dmean speed (fps): 0.38 -> 0.51Improved
train on single pigeon, test on 5 pigeons instead of 1 (ablation)KeypointRCNN · 3D-POP 2DPCK10 (%): 98.7 -> 83.6Hurt
train on single pigeon, test on 10 pigeons instead of 1 (ablation)KeypointRCNN · 3D-POP 3Dmedian error (mm): 6.9 -> 246.7Hurt
fine-tune DLC from indoor weights on wild data instead of indoor-only training (ablation)Wild-DLC · Wild-MuPPET 3DPCK10 (%): 74.4 -> 81.3Improved
train DLC from scratch on wild data instead of fine-tuning (ablation)DLC-Fine-tuned · Wild-MuPPET 3Dmedian error (mm): 11.4 -> 12.7No 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