Go to Alexandria's home page
The Library of Alexandria

Machine vision: 2D collages

Jim Carnicelli's AI Blog

Alexandria Home | Up One Level

ò
Switch to multi-page mode for smaller pages with cross-navigation.    Switch to single-page mode for all content in one page.

Thursday, 6/9/2005

Machine vision: 2D collages

Back to
blog home
Listen to an
audio version
Notify me of
new entries
Subscribe to a full
RSS feed of this blog

Following is another in my series of ad hoc journal entries I've been keeping of my thoughts on machine vision.

I've been nursing the idea that it's not necessary to have a detailed sense of how far away things in an image are. It's probably sufficient, in some basic contexts, to just know that one thing is in front of another and not care about absolute distances. It seems some MV researchers have gone ape over telling exactly how far away an apple on a table is using lasers, stereo displacement, and all sorts of tricks. Maybe just knowing how big an apple typically is is good enough for telling how far away it is.

When I think about 3D vision in this context, I have been likening the visible world to a collage of 2D images. Take the scene seen by a stationary camera looking at a road as cars go by. One could take the unchanging background as one image. A car moving by would be the only object of interest. What's interesting is that the image of the car, from snapshot to snapshot, doesn't change much. It's as though one just took the previous image of the car and stretched and warped it a little in order to get the current image of the car. That "smooth morphing" idea is at the heart of this 2D collage analogy.

In the car example, it should be fairly easy to use the conventional technique of seeing pixel differences between a before and after image to isolate the car from the background. Not sure yet how to deal with the morphing. It seems, fair, though, to assume that the car doesn't just disappear unless it's heading out of the scene. Instead, it should suffice to take the "before car" and place it in the "after car" space and then scale it to fit the blob. Then comes a comparison step to see how the two car images differ. Perhaps key points - edges or corners - can be found and their positions corresponded.

method="post" action="../../ai/feedback.asp">
Your Feedback
Name (optional):
Email (optional):

Prove Your Humanity:
Please enter the code you see here. This is designed to
protect our message board from spam posted by automated software.
Those programs can't easily read these codes like you and I can.

Subject: AI - Blog - Machine vision: 2D collages
Or write me an email instead.         

Back to
blog home
Listen to an
audio version
Notify me of
new entries
Subscribe to a full
RSS feed of this blog


All Entries

(reverse date order)

  • 11/13/2007 - Confirmation bias as a tool of perception
  • 11/6/2007 - What bar code scanners can tell us about perception
  • 10/21/2007 - Perception as construction of stable interpretations
  • 10/14/2007 - Rebuttal of the Chinese Room Argument
  • 10/7/2007 - Video stabilizer
  • 9/27/2007 - "Conscious Realism" and "Multimodal User Interface" theories
  • 7/4/2007 - Plan for video patch analysis study
  • 7/1/2007 - Patch mapping in video
  • 6/27/2007 - Emotional and moral tagging of percepts and concepts
  • 6/22/2007 - A hypothetical blob-based vision system
  • 4/21/2007 - Abstraction in neuron banks
  • 4/12/2007 - Pattern Sniffer: a demonstration of neural learning
  • 4/7/2007 - A respectful critique of the Hierarchical Temporal Memory (HTM) concept
  • 11/10/2005 - Neuron banks and learning
  • 11/3/2005 - A standardized test of perceptual capability
  • 10/29/2005 - Using your face and a webcam to control a computer
  • 10/8/2005 - Stereo disparity edge maps
  • 9/25/2005 - Some stereo vision illusions
  • 9/21/2005 - Topics in Machine Vision
  • 8/26/2005 - Introduction to Machine Vision
  • 8/14/2005 - Bob Mottram, crafty fellow
  • 8/11/2005 - Stereo vision: measuring object distance using pixel offset
  • 8/7/2005 - Automatic alignment of stereo cameras
  • 8/7/2005 - DualCameras component
  • 7/30/2005 - Patch equivalence
  • 7/12/2005 - Machine vision: motion-based segmentation
  • 6/20/2005 - Machine vision: spindles
  • 6/16/2005 - Machine vision: smoothing out textures
  • 6/15/2005 - Machine vision: studying surface textures
  • 6/10/2005 - Machine vision: pixel morphing
  • 6/10/2005 - Machine vision: motion tracking
  • 6/10/2005 - Machine vision: tilting my head
  • 6/10/2005 - Machine vision: layer-based models
  • 6/9/2005 - Machine vision: 2D collages
  • 6/9/2005 - Machine vision: Hierarchy of regions
  • 6/9/2005 - Machine vision: cost-effective action
  • 6/9/2005 - Machine vision: overlooking shadow and light splotches on surfaces
  • 6/9/2005 - Machine vision: blob growth
  • 5/11/2005 - Review of "Visual Intelligence"
  • 5/4/2005 - The portable, hand-held learning laboratory
  • 4/27/2005 - Review of "On Intelligence"
  • 4/15/2005 - Bubble Vision
  • 2/26/2005 - Machine vision of GUIs
  • 1/23/2005 - The fallacy of bigger brains
  • 1/12/2005 - Follow-up on Pile
  • 1/12/2005 - A review of the premises behind Pile
  • 11/28/2004 - Thoughts on FLARE
  • 11/28/2004 - New project: Mechasphere
  • 11/14/2004 - Review of "Bicentennial Man"
  • 11/2/2004 - Neural network demo
  • 10/17/2004 - Roamer: recent updates
  • 10/13/2004 - New Roamer project
  • 10/9/2004 - First entry


    Go to Alexandria's home page Copyright © 2010 The Library of Alexandria. All rights reserved.
    Produced in cooperation with Carnell Information Systems, Inc.