Binocular Stereoscopic Vision Notes

Binocular Stereoscopic Vision

Binocular vision is the main source of three-dimensional perception, enabling us to perceive depth and spatial relationships.
This lecture will further explore how our two eyes, positioned side-by-side, synergistically contribute to depth perception. We'll use red-green anaglyphs to simulate 3D vision, providing a hands-on demonstration of the principles involved.

Retinal Disparity

Retinal disparity refers to the slight horizontal positional difference in the images detected by each eye. This difference is the foundation of our stereoscopic vision.
This disparity arises due to the physical separation, or distance, between our two eyes, each capturing a slightly different perspective of the same scene.
Retinal disparity is a critical source of three-dimensional depth perception, allowing us to perceive the depth and distance of objects accurately.
Red-green glasses are used to simulate and mimic retinal disparity on a two-dimensional image. These glasses work by filtering images with a slight horizontal shift, where each eye sees a slightly different image.
This principle is foundational to 3D movies and virtual reality technologies, which create an immersive experience by manipulating retinal disparity.

Eye Arrangement in Different Species

Different species exhibit diverse eye arrangements, each influencing the field of view and the capacity for stereoscopic vision.
Reptiles typically feature eyes positioned side-by-side but with a greater interocular distance compared to humans, impacting their depth perception and field of view.
Fish generally have eyes located on the sides of their heads, affording them a wider field of view but often at the cost of limited depth perception.
Humans and cats possess eye arrangements that optimize three-dimensional depth perception, enabling them to navigate and interact with their environment effectively.

Vergence Angle

Vergence angle is defined as the angle formed by the two eyes when converging on a single object. This angle is crucial for binocular vision.
The vergence angle varies depending on the object's distance. Larger angles indicate closer objects, while smaller angles suggest more distant ones.
This angle provides essential information about absolute distance and is a key component of binocular vision, contributing to our sense of depth.

Accommodation

Accommodation involves the modulation of the lens thickness in our eyes, allowing us to focus on objects at varying distances. This process is vital for clear vision.
By changing the refractive power of the lens, accommodation ensures that images are sharply focused on the retina, regardless of object distance.
Accommodation is another significant source of binocular vision, working in tandem with vergence angle to provide us with a comprehensive sense of depth and spatial awareness.

Wheatstone's Stereoscope

Sir Charles Wheatstone invented the stereoscope as a tool to investigate retinal disparity and its significance in three-dimensional perception.
This device employs two mirrors positioned at a 90-degree angle, enabling the presentation of distinct images to each eye independently.
By presenting stimulus pairs with subtle horizontal positional shifts, Wheatstone's stereoscope effectively mimics retinal disparity, creating the illusion of depth.
As a result, it allows for the perception of three-dimensional depth from two-dimensional pictures, offering valuable insights into how our visual system processes depth information.

Discomfort in Virtual Reality

Discomfort or nausea experienced during virtual reality or while viewing 3D content can stem from a conflict between retinal disparity and accommodation.
In virtual reality, retinal disparity is artificially induced, yet natural accommodation is absent because the images are displayed on a two-dimensional screen.
This mismatch can lead to symptoms such as dizziness or discomfort, with sensitivity varying among individuals, highlighting the challenges of creating truly immersive virtual experiences.

Anaglyphs

Anaglyphs are images crafted using red and green filters to simulate the effect of retinal disparity.
This is achieved by capturing an image and generating a slightly horizontally shifted version, with each image then filtered through either a red or green lens before being superimposed.
When viewed through red-green glasses, anaglyphs enable observers to perceive three-dimensional depth, illustrating a practical application of retinal disparity manipulation.

Stereopsis and Disparity

Stereopsis denotes the three-dimensional vision that arises from subtle differences between the images seen by the left and right eyes.
Disparity refers to the slight variations in the positions of features in the views of the left and right eyes.
Zero disparity occurs when the focus point and the objects are situated in the same plane, indicating no relative depth difference.
Cross disparity arises when an object is situated closer to the viewer than the fixation point, signaling a nearer depth.
Uncross disparity occurs when the object is positioned farther away from the viewer than the fixation point, indicating a greater depth.
The magnitude of disparity serves as an indicator of the extent of depth difference between objects.

Demonstrations of Disparity

Zero Disparity: Demonstrated by positioning two fingers in the same depth plane, ensuring no distance difference between them.
Uncross Disparity: Achieved by using one finger as an anchor and placing the other finger farther away from the viewer. This results in the distance between the fingers appearing narrower when viewed with the left eye compared to the right eye.
Cross Disparity: Demonstrated by using one finger as an anchor and placing the other finger closer to the viewer. This causes the distance between the fingers to appear narrower when viewed with the left eye compared to the right eye.
The magnitude of disparity provides valuable information about the relative depth difference between objects, enhancing our depth perception.

Disparity Varies

Disparity magnitude is influenced by the viewing distance, affecting our stereoscopic perception.
Stereopsis is most effective within a range of 10 to 20 feet from the observer, beyond which depth perception diminishes.
The magnitude of disparity is also contingent on interpupillary distance, which varies among individuals.

Stereo Acuity

Stereo acuity is defined as the smallest disparity that can be resolved by an observer, indicating the precision of depth perception.
Under optimal conditions, human stereo acuity is approximately five arc seconds, highlighting our remarkable ability to discern subtle depth differences.
(1 \, \text{degree of visual angle} = 60 \, \text{arc minutes})
(1 \, \text{arc minute} = 60 \, \text{arc seconds})

Random Dot Stereograms

Random dot stereograms were devised by Bela Julesz in 1971 and represent a groundbreaking approach to studying depth perception.
These stereograms consist of black or white dots randomly arranged within a rectangular space, devoid of any recognizable shapes or forms.
They provide stereo disparity without relying on any other visual cues, isolating the effect of disparity on depth perception.
In their construction, a shape in the middle of one random dot image is deliberately shifted to the right or left by a small amount, and the resulting empty space is filled with additional random dots.
Despite each image appearing largely similar when viewed separately, the introduction of stereo disparity creates a compelling sensation of three-dimensionality when viewed together. These stimuli are invaluable for assessing stereo vision independent of other visual information.

The Brain and Retinal Disparity

The brain undertakes the task of identifying corresponding features in each eye's view to facilitate matching and depth perception.
This is achieved by maximizing matches between the two images, enabling the brain to effectively process binocular information.
Certain cells are specialized to respond to zero disparity, indicating objects in the same depth plane, while others are tuned to cross disparity or uncross disparity, signaling relative depth differences.
Neurons within the visual cortex play a crucial role in matching features between the two eyes and computing retinal disparity, enabling depth perception.
Disparity information needs to be scaled based on distance, necessitating the integration of non-stereoscopic cues such as monocular depth cues and physiological measures like body movement and motion information, to achieve accurate stereoscopic scaling.

Binocular Rivalry

Binocular rivalry arises when markedly dissimilar images are presented to each eye simultaneously, leading to competition for visual dominance.
This competition results in each image being perceived for brief intervals, alternating spontaneously as the visual system grapples with conflicting input.
Binocular rivalry serves as a specific instance of bistable vision, where perception oscillates between two distinct interpretations of the same stimulus.
In a brain imaging study conducted by Frank Tong, participants were presented with a face to one eye and a house to the other eye, eliciting binocular rivalry.
During binocular rivalry, despite the constant stimuli, perception alternates between the face and the house, reflecting the dynamic nature of visual processing.
Notably, the ventral stream areas, including the FFA (fusiform face area) and PPA (parahippocampal place area), exhibited activity patterns that paralleled perceptual changes rather than directly mirroring the physical characteristics of the stimuli.