Advancing pig lameness detection with a multi-task framework: Leveraging multi-view 3D pose estimation and wavelet convolution

Li Xiang

doi:10.1016/j.biosystemseng.2026.104393

Advancing pig lameness detection with a multi-task framework: Leveraging multi-view 3D pose estimation and wavelet convolution

2026-01-15

0

PRE

AI

DOI：10.1016/j.biosystemseng.2026.104393

Original

Original request for help

Save

Abstract

En 中文

• Multi-view 3D pose estimation reconstructs pig skeletons for lameness detection. • Low-dimensional 3D gait and motion features for better spatial-temporal analysis. • The PoseGait-MT framework detects lameness severity and identifies affected limbs. • WTConv stabilises training for smoother convergence and enhanced feature learning.

Advancing pig lameness detection with a multi-task framework: Leveraging multi-view 3D pose estimation and wavelet convolution

Abstract

Journal

Organization