High-Resolution Light Field Capture With Coded Aperture.

Area of Science:

• Optics and Photonics
• Computational Imaging
• Computer Vision

Background:

• Light field capture aims for high angular and spatial resolution cost-effectively.
• Traditional methods often involve trade-offs between angular and spatial resolution.
• Coded apertures offer a way to multiplex views, potentially overcoming resolution limitations.

Purpose of the Study:

• To propose an improved light field camera model.
• To enhance light field reconstruction quality and light efficiency.
• To avoid the typical trade-off between angular and spatial resolution in light field capture.

Main Methods:

• Developed a light field camera model with double compressive measurements and a single mask.
• Implemented compressive measurements using a coded aperture and a random convolution CMOS imager.
• Utilized a single mask design for high light efficiency and sampling clarity.

Main Results:

• The proposed model achieved higher reconstruction quality compared to representative mask-based light field cameras.
• Demonstrated higher light efficiency due to the single mask design.
• The double measurement strategy preserved more correlation information, improving the reconstructed light field.

Conclusions:

• The improved light field camera design effectively enhances reconstruction quality and light efficiency.
• The proposed method successfully avoids the resolution trade-off inherent in many light field capture systems.
• This approach offers a promising direction for high-performance, cost-effective light field acquisition.