Polarization-based approach for multipath interference mitigation in time-of-flight imaging.

Area of Science:

• Optics and Photonics
• Computer Vision
• Robotics

Background:

• Time-of-flight (ToF) cameras are crucial for depth sensing but suffer from errors like multipath interference due to nearby obstructions.
• Multipath interference occurs when light signals reflect off multiple surfaces before reaching the sensor, distorting depth measurements.

Purpose of the Study:

• To develop and validate a polarized time-of-flight system for mitigating multipath interference in depth sensing.
• To enhance the accuracy of depth maps in the presence of specular obstructions.

Main Methods:

• A polarized time-of-flight imaging system was designed and implemented.
• The system leverages polarization cues and phasor representation of ToF data.
• Experiments were conducted with various targets (rough and smooth) and obstructions.

Main Results:

• The proposed method successfully acquires depth maps with high accuracy, even with specular obstructions.
• The system demonstrates effectiveness for both rough and smooth surfaces, which have different polarization characteristics.
• Experimental results qualitatively and quantitatively confirm the method's performance.

Conclusions:

• Polarization-based ToF imaging offers a robust solution for multipath interference mitigation.
• The developed system significantly improves depth sensing accuracy in challenging environments with obstructions.
• This approach broadens the applicability of ToF cameras in scenarios with complex reflective surfaces.