Project Abstract

MotivationThe research aims to improve 3D human pose estimation accuracy using artificial intelligence. It explores the importance of lightweight networks and specialized datasets in developing AI-based models. Technologies used include CNN, GCN, Deep Kinematic pose regression, Twin Gaussian Processes, LSTM, and TCNs. The Texel Portal 3D Scanner is used for data collection. The study focuses on balancing computational effectiveness with modelling precision, using Deep Physics Modelling for Unilaterally 3D Human Posture Estimation. The project aims to understand the advantages of AI models and spatial data in 3D human pose estimation.Research questionWhat is the importance of developing lightweight networks for 3D human pose estimation?Why is the use of specialized datasets crucial in developing AI-based models for 3D human pose estimation?Methods employedGraph Convolutional Neural Networks (GCNs) are being used in AI for 3D Human Pose Analysis to accurately represent body connections, demonstrating promising capabilities for 3D human pose estimation.Main FindingsThe study aims to improve 3D human pose estimation accuracy, durability, and practical use of AI models. It focuses on strengthening model extension, overcoming occlusion issues, balancing precision and effectiveness, and optimizing spatial data usage. The goal is to create AI systems that accurately forecast human body positions, enabling applications in augmented reality, animation, athletics evaluation, healthcare, and independent nations systems.

Keywords: Algorithms, Data Visualisation, Computer Vision

 Conference Details

Session: Presentation Stream 23 at Presentation Slot 7

Location: College 127 at Wednesday 8th 09:00 – 12:30

Markers: Eike Neumann, Julian Hough

Course: MSc Advanced Computer Science, Masters PG

Future Plans: I’m looking for an industry placement