Perception-based 3D tactile rendering from a single image for human skin examinations by dynamic touch

Area of Science:

• Biomedical Engineering
• Haptics
• Dermatology

Background:

• Skin condition diagnosis relies heavily on tactile properties like stiffness and friction, not just visual cues.
• Integrating tactile feedback into vision-based systems can enhance diagnostic accuracy for dermatologists.

Purpose of the Study:

• To develop a tactile rendering system for real-time skin examinations using dynamic touch.
• To create a system that translates 2D skin images into 3D haptic surfaces.

Main Methods:

• The system converts single 2D skin images into 3D haptic surfaces, informed by human psychophysical data on skin perception.
• Real skin biomechanical properties were incorporated into the rendering process.
• The system is designed as a standalone unit compatible with various cameras and haptic devices.

Main Results:

• An identification experiment with five subjects demonstrated the system's ability to render discernible tactile differences between skin surfaces.
• Participants successfully identified different skin surfaces using only the haptic device, without visual input.

Conclusions:

• The developed system generates 3D haptic surfaces from single skin images, based on human haptic perception.
• It offers realistic, real-time skin interactions for diagnosis, simulation, and training.
• Potential applications extend to virtual reality and cosmetic industries.