Motion Capture Laboratory
The Specialist Laboratory for Vision Intelligence focuses on interdisciplinary research combining computer vision, machine learning, and deep learning. Research areas include:
- Analysis of three-dimensional motion
- Human activity recognition
- Human emotion recognition
- Detection of incorrect actions
- Extraction and understanding of information in static images and video recordings (e.g., content-based image retrieval)
- 3D model and animation creation
The laboratory is equipped with advanced hardware for interdisciplinary research:
- Passive optical Vicon motion capture system
- Data storage servers
Vicon Motion Capture System
The system consists of:
- 8 Vicon T40S near-infrared cameras
- Two Bonita 720c reference video cameras
- Giganet hub
- Myon 320 EMG system
- Two AMTI biomechanical platforms
Vicon T40S near-infrared cameras combined with Bonita 720c video cameras precisely track object movement in 3D. The T40S cameras capture high-resolution (4-megapixel) images at up to 515 frames per second, enabling accurate reproduction of even very fast-moving objects. Bonita cameras are used to observe details of the examined objects. The wireless EMG system allows for recording 16 channels of electromyographic signals synchronized with motion data, significantly increasing comfort for the examined person. Biomechanical platforms precisely measure ground reaction forces up to 17,000N. The included software enables modeling, processing motion data, creating reports, and presenting results.
The motion capture laboratory allows for a wide range of motion-focused research, which can be combined with electromyographic and biomechanical studies.
The laboratory serves both research and education purposes. Advanced research conducted as part of engineering and master’s theses enhances students’ education and enhances the skills of graduates, giving them a better start in their careers
PhaseSpase Motion Capture System
The system consists of:
- 28 PhaseSpace Impulse X2 cameras
- Server with hub and base station
- Active LED markers
- 15 LED microdrivers
- 8 Full body motion capture suits
The PhaseSpace motion capture system is a sophisticated solution designed for precise tracking of three-dimensional object movement. Equipped with high-resolution (3600×3600) cameras, the system provides detailed motion data essential for various applications, including virtual reality and sports motion analysis. Utilizing active LED technology, each marker is assigned a unique ID, ensuring accurate tracking and eliminating the need for marker-swapping. The system offers real-time tracking at a speed of 960 frames per second with a latency of less than 3 milliseconds. This enables smooth and accurate recording of even the most rapidly changing scenes. PhaseSpace’s API/SDK provides support for third-party applications, facilitating real-time user interaction and seamless integration with existing software tools.