29-06-2012, 12:40 PM
Sight in an Unstructured World – 3D Stereo Vision and Systems
[attachment=26071]
Abstract
For decades, the business world has sought a practical way to add
‘computer vision’ capabilities to applications and products. Industries
understand the windfall of new business benefits that computer vision
could introduce – from more sophisticated homeland defense
capabilities, to improvements in automotive safety, to smarter robots,
to more interactive video games, and more.
This paper describes the challenges for prior approaches to computer
vision, and highlights 3D stereo vision’s emergence as the preferred
method of enabling sight in machines.
Introduction
Of the five human senses, sight is generally regarded as the most
important for exploring and understanding the world around us. So it’s
no surprise that the business world is excited about using computer
vision to enable systems to better see, interpret and respond to real
world events, in real-time.
However, despite the business community’s strong desire to find
practical ways to incorporate computer vision into commercial offerings,
early computer vision approaches fell short. The few examples that
worked well were limited to tightly controlled research lab and
manufacturing environments. They have not proven practical for
real-world deployment.
Using Cameras to See – Historic Limitations
The scientific community has always used cameras as
eyes for machine vision. However, for a variety of technical
reasons, it is a challenge for computers to interpret cameragenerated
images in the same manner as humans would.
The real world is in constant motion, full of shifting backgrounds
and objects of different shapes and sizes. The ability
to interpret this dynamic landscape is computationally and
theoretically problematic. When a computer looks at the
world through a lens it just sees pixels. A pixel doesn’t
make sense on its own. While image sensors (driven by
Moore’s Law and the digital camera market) continue to make
dramatic improvements in cost and resolution, the abilities to
interpret visual data and to do so in real-time have been the
missing links.
Why 3D Vision is Better than 2D
A major breakthrough in computer vision occurred when
researchers turned to biology and the study of stereo vision
to turn 2D images into 3D volumes. This approach better
represents distance, size and spatial relationships between
different objects in the camera’s field of view.
For example, the image of the woman and the child changes
dramatically when viewed in 3D from stereo vision (darker
pixels are closer).
What about Lidar and Radar?
In addition to 2D vision, there have also been attempts to realize
the promise of systems that see by using Radar and Lidar.
Radar is the method of detecting distant objects and
determining their position, velocity, or other characteristics
by analysis of very high frequency radio waves reflected
from their surfaces. Lidar is the method of detecting distant
objects and determining their position, velocity, or other
characteristics by analysis of pulsed laser light reflected
from their surfaces.
Radar and Lidar have shortcomings when it comes to the
industry’s need for systems that see. Both approaches are
‘active’ – meaning they rely on broadcasting and returning
echoes (RF and light, respectively). In both cases, this need
to send/receive a signal tends to have a very negative impact
on the accuracy and resolution of the 3D results generated.
They tend to be very coarse measurements (particularly as
one gets further away from the sensor),
[attachment=26071]
Abstract
For decades, the business world has sought a practical way to add
‘computer vision’ capabilities to applications and products. Industries
understand the windfall of new business benefits that computer vision
could introduce – from more sophisticated homeland defense
capabilities, to improvements in automotive safety, to smarter robots,
to more interactive video games, and more.
This paper describes the challenges for prior approaches to computer
vision, and highlights 3D stereo vision’s emergence as the preferred
method of enabling sight in machines.
Introduction
Of the five human senses, sight is generally regarded as the most
important for exploring and understanding the world around us. So it’s
no surprise that the business world is excited about using computer
vision to enable systems to better see, interpret and respond to real
world events, in real-time.
However, despite the business community’s strong desire to find
practical ways to incorporate computer vision into commercial offerings,
early computer vision approaches fell short. The few examples that
worked well were limited to tightly controlled research lab and
manufacturing environments. They have not proven practical for
real-world deployment.
Using Cameras to See – Historic Limitations
The scientific community has always used cameras as
eyes for machine vision. However, for a variety of technical
reasons, it is a challenge for computers to interpret cameragenerated
images in the same manner as humans would.
The real world is in constant motion, full of shifting backgrounds
and objects of different shapes and sizes. The ability
to interpret this dynamic landscape is computationally and
theoretically problematic. When a computer looks at the
world through a lens it just sees pixels. A pixel doesn’t
make sense on its own. While image sensors (driven by
Moore’s Law and the digital camera market) continue to make
dramatic improvements in cost and resolution, the abilities to
interpret visual data and to do so in real-time have been the
missing links.
Why 3D Vision is Better than 2D
A major breakthrough in computer vision occurred when
researchers turned to biology and the study of stereo vision
to turn 2D images into 3D volumes. This approach better
represents distance, size and spatial relationships between
different objects in the camera’s field of view.
For example, the image of the woman and the child changes
dramatically when viewed in 3D from stereo vision (darker
pixels are closer).
What about Lidar and Radar?
In addition to 2D vision, there have also been attempts to realize
the promise of systems that see by using Radar and Lidar.
Radar is the method of detecting distant objects and
determining their position, velocity, or other characteristics
by analysis of very high frequency radio waves reflected
from their surfaces. Lidar is the method of detecting distant
objects and determining their position, velocity, or other
characteristics by analysis of pulsed laser light reflected
from their surfaces.
Radar and Lidar have shortcomings when it comes to the
industry’s need for systems that see. Both approaches are
‘active’ – meaning they rely on broadcasting and returning
echoes (RF and light, respectively). In both cases, this need
to send/receive a signal tends to have a very negative impact
on the accuracy and resolution of the 3D results generated.
They tend to be very coarse measurements (particularly as
one gets further away from the sensor),