Optimal allocation of quantized human eye depth perception for multi-focal 3D display design

Area of Science:

• Optics and Photonics
• Human-Computer Interaction
• Visual Perception

Background:

• Advancements in 3D stereoscopic, autostereoscopic, and lightfield displays are central to head-mounted and lightfield display optical design.
• A lack of consensus exists regarding the necessary quantized depth levels for these emerging displays in stereoscopic and monocular modes.

Purpose of the Study:

• To define and prioritize quantized depth levels that saturate human depth perception based on psychophysical theories.
• To develop an optimization framework for globally locating depth levels in band-limited displays.

Main Methods:

• Utilized psychophysical theories to guide the definition of depth levels.
• Developed a general optimization framework for band-limited displays.
• Reformulated an intractable problem into a tractable one involving maximally covering regions with hypographs representing monocular depth of field.

Main Results:

• An average of 1731 stereoscopic and 7 monocular depth levels are required to saturate visual depth perception from 25 cm to infinity.
• Optimal allocation of the first three monocular depth levels is critical for minimizing population-wide error.
• The study details the 3D spatial profile of quantized stereoscopic and monocular depth levels.

Conclusions:

• The findings provide fundamental guidelines for designing optimal near-eye displays, lightfield monitors, and 3D screens.
• Further increases in depth levels beyond the identified optima yield negligible improvements in visual perception.