AI Paper Atlas

Research map / 3D vision / Human pose estimation

MooCap: A Multi-View Benchmark for Cow-Object-Human Interaction and Behavior Dynamics

Ian Noronha, Heather Neave, Upinder Kaur

CVPR 2026 · 3D vision › Human pose estimation · approach: Temporal video pose estimation

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 hierarchical Transformer FACT instead of diffusion-based DiffAct for long action segmentationtemporal action segmentation · MooCapAcc (MoF): 35.65 -> 66.39Improved
use long-range temporal modeling (LTContext) over ASFormertemporal action segmentation · MooCapAcc (MoF): 34.15 -> 48.87Improved
use unsupervised clustering instead of supervised segmentationtemporal action segmentation · MooCapAcc (MoF): 66.39 -> 14.73Hurt
use MS-G3D instead of 2S-AGCN for pose-based recognitionskeleton behavior recognition · MooCapmean F1: 0.26 -> 0.39Improved
use UniFormer instead of other video transformers for rearing-group classificationlongitudinal behavioral classification · MooCap, mean of 7 scenariosaccuracy (%): 45.50 -> 84.23Improved
switch scenario from dam reunion to human approachTimeSformer rearing-group classification · MooCapaccuracy (%): 96.67 -> 22.22Hurt
use skeleton input alone for close-proximity recognitionMS-G3D · MooCapF1: 0.50 (Close Proximity, MS-G3D) vs 0.02 (Attentive)Mixed

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