Perceptual Quality Assessment of Compressed Vibrotactile Signals Through Comparative Judgment.

Area of Science:

• Human-Computer Interaction
• Signal Processing
• Psychophysics

Background:

• Assessing the quality of compressed vibrotactile signals is crucial for haptic technology development.
• Existing methods may lack standardization or introduce experimental confounds.
• The Multiple Stimulus Test with Hidden Reference and Anchors (MUSHRA) is a validated method in audio quality assessment.

Purpose of the Study:

• To present a novel, comprehensive, and validated scheme for the quality assessment of compressed vibrotactile signals.
• To adapt and refine the MUSHRA methodology for vibrotactile signal evaluation.
• To provide a standardized and reliable method for comparing the performance of vibrotactile codecs.

Main Methods:

• A hierarchically structured and strictly timed assessment procedure was designed, inspired by MUSHRA.
• Human assessors compared compressed vibrotactile signals against original, hidden reference, and anchor signals.
• Numerical ratings were collected, with hidden references and anchors used for validation and assessor screening.

Main Results:

• The proposed method effectively enabled the comparison of performance between two state-of-the-art lossy vibrotactile codecs.
• Subjective quality was reliably measured across different data compression rates.
• The experiment with 20 participants validated the efficacy and precision of the assessment scheme.

Conclusions:

• The developed method offers a robust and adaptable solution for evaluating vibrotactile codec quality.
• This approach minimizes experimental confounds and provides precise psychophysical assessments.
• The scheme can be readily applied to assess other vibrotactile codecs and compression techniques.