DIGITAL AVATARS AS A MEANS OF COMPARING MARKER- BASED MOTION CAPTURE SYSTEMS AND ARTIFICIAL INTELLIGENCE-BASED MOTION CAPTURE SYSTEMS

• Main Author: Aditya Samiadji, Wisnu
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/75266
• Institution: Institut Teknologi Bandung

With the advancement of Artificial Intelligence (AI) technology, the production cost and complexity of creating a motion capture system have become more affordable, simpler, and accessible to small content creation industries. However, due to the absence of markers in the motion estimation process, and only relying on estimated key points of the actor's visible body parts, the accuracy of AI-based motion capture systems can be a concern for industries seeking accurate results without the high costs associated with conventional motion capture systems. To address this issue, a cost-effective and simple version of the conventional motion capture system is developed, utilizing color-based markers and multiple cameras to capture actor movements. The focus of this final project is to integrate the simple marker-based and pose-based motion capture systems with digital avatar media, which will be used for comparison between the two systems. To achieve this goal, the final application is developed using Blender for digital avatar creation and Unity to simulate avatar movements based on the key points obtained from each motion capture system. The avatar's body movements and rotations are designed to adhere to human limits and leverage articulated object principles. The testing results show that the color-based marker motion capture system exhibits better movement than the pose-based system. However, the developed system still has limitations, such as the restricted number of markers and colors that can be used as key points in the color-based marker motion capture system.