Abstract:Motion parameters measurement is essential for understanding animal behavior, exploring the laws of object motion, and studying control methods. Nowadays, advanced computer vision based on machine learning technology supports markerless object tracking in 2D videos. However, due to the fact that all objects move in three-dimensional space, this paper introduces a method of measuring motion parameters using 3D pose estimation. First, an enhanced iterative bundle adjustment algorithm is proposed for multi-camera calibration in a multi-camera vision system by adding two control parameters, which dramatically reduces the reprojection error of multi-camera calibration and lays the foundation for high-precision triangulation. Then, a new spatiotemporal loss function is proposed, which considers the relationship between key points that do not constitute limbs, thereby improving triangulation accuracy. The new multi-camera calibration algorithm is evaluated on ChArUco and 3D pose estimation for metronome, planet pendulum, human hand, Koi, and cheetah. The experimental results show that: (1) the two hyper-parameters in the enhanced iterative bundle adjustment algorithm effectively suppress the influence of noise and play a good role in reducing the reprojection error of multi-camera calibration; (2) the spatiotemporal loss function has a strong constraining ability, the time loss can stabilize high frame rate video triangulation to maintain accuracy, while the space loss can improve the accuracy of triangulation for more complex structures; (3) multi-view data fusion is also conducive to improving the accuracy of triangulation. Moreover, the method was successfully applied to some actual measurement scenes: (1) the accurate measurement of the frequency of a metronome; and (2) the success measurement of the movement of a Koi, which conforms to the basic model of fish swimming. Some dynamic measurement results are displayed at https://github.com/wux024/AdamPose .

Towards Multi-Modal Animal Pose Estimation: An In-Depth Analysis

Motion Parameters Measurement of User-Defined Key Points Using 3D Pose Estimation

AP-10K: A Benchmark for Animal Pose Estimation in the Wild

SemiMultiPose: A Semi-supervised Multi-animal Pose Estimation Framework

A Comprehensive Survey on 2D Multi-Person Pose Estimation Methods

Animal3D: A Comprehensive Dataset of 3D Animal Pose and Shape

The Overview of Multi-person Pose Estimation Method

PMotion: An advanced markerless pose estimation approach based on novel deep learning framework used to reveal neurobehavior

SuperAnimal pretrained pose estimation models for behavioral analysis

Pose for Everything: Towards Category-Agnostic Pose Estimation

Multi-Person Full Body Pose Estimation

MPE-HRNetL: A Lightweight High-Resolution Network for Multispecies Animal Pose Estimation

A Real-time Multi-Subject Three Dimensional Pose Tracking System for Analyzing Social Behaviors of Non-human Primates

AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time

Multi-animal 3D Social Pose Estimation, Identification and Behaviour Embedding with a Few-Shot Learning Framework

Look Into Multi-Person: A New Benchmark For Pose Estimation And Human Parsing

Of Mice and Pose: 2D Mouse Pose Estimation from Unlabelled Data and Synthetic Prior

Articulated Pose Estimation with Global Body Orientation

APT-36K: A Large-scale Benchmark for Animal Pose Estimation and Tracking

EAPoseNet: efficient animal pose network in low computing power scenarios

Multi-animal pose estimation, identification and tracking with DeepLabCut