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 .

MPEG-4 Parameter Computing with Multi-curve Spectrum of Eyes

Learning and Synthesizing Mpeg-4 Compatible 3-D Face Animation from Video Sequence

Animating 3D facial models with MPEG-4 FaceDefTables

MPEG-4 Based Facial Expression Image Morphing

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

MPEG-4 Facial Animation Techniques and the Design of a MPEG-4 based Facial Animation System

Video-Based Facial Animation with Detailed Appearance Texture

Animated Face Modeling and Adaptation with Triangular B-Spline

Speech Driven MPEG-4 Based Face Animation via Neural Network

Facial Expression Animation Based on Physical Model

Video Tracked Facial Expression Animation

Animating Arbitrary Topology 3D Facial Model Using the MPEG-4 FaceDefTables

Personalized facial animation based on 3d model fitting from two orthogonal face images

Realistic 3D Facial Animations with Partial Expression Ratio Image

Real-time Synthesis of Chinese Visual Speech and Facial Expressions Using MPEG-4 FAP Features in a Three-Dimensional Avatar

Modeling the Correlation Between Modality Semantics and Facial Expressions

Real-time Synthesis of Chinese Visua using MPEG-4 FAP Features in a

Performance-Driven Animation of Hand-Drawn Cartoon Faces

An Efficient Approach for Multi-View Face Animation Based on Quasi 3d Model

Quick Customization of Particular Human Face and Muscle-Driven Expression Animation

CurveMEF: Multi-eposure fusion via curve embedding network