SuperMag: Vision-based Tactile Data Guided High-resolution Tactile Shape Reconstruction for Magnetic Tactile Sensors

-- Magnetic-based tactile sensors (MBTS) combine the advantages of compact design and high-frequency operation but suffer from limited spatial resolution due to their sparse taxel arrays. This paper proposes SuperMag, a tactile shape reconstruction method that addresses this limitation by leveraging high-resolution vision-based tactile sensor (VBTS) data to supervise MBTS super-resolution. Co-designed, open-source VBTS and MBTS with identical contact modules enable synchronized data collection of high-resolution shapes and magnetic signals via a symmetric calibration setup. The MBTS achieves a sampling frequency of 125 Hz, whereas the shape reconstruction sustains an inference time within 2.5 ms. Tactile sensing is essential in robotics, enabling agents to perceive and interact with their environment through physical contact [1, 2]. Inspired by the biological sense of touch, tactile sensors detect mechanical stimuli such as contact force, texture, slip, and vibrations. Common sensing technologies include capacitive [3], resistive [4], piezoresistive [5], piezoelectric [6], triboelectric [7], barometric [8], optical [9], and magnetic [10] sensors, each offering unique advantages for tactile perception and robotic applications.