Classifying Objects from Visual Information
Consider a world of 'objects.' Our goal is to place these objects into categories that are useful to the observer using sensory data. One criterion for utility is that the categories allow the observer to infer the object's potential behaviors, which are often non-observable. Und...
Main Authors: | , |
---|---|
Language: | en_US |
Published: |
2004
|
Online Access: | http://hdl.handle.net/1721.1/6443 |
_version_ | 1811073050675773440 |
---|---|
author | Bobick, Aaron Richards, Whitman |
author_facet | Bobick, Aaron Richards, Whitman |
author_sort | Bobick, Aaron |
collection | MIT |
description | Consider a world of 'objects.' Our goal is to place these objects into categories that are useful to the observer using sensory data. One criterion for utility is that the categories allow the observer to infer the object's potential behaviors, which are often non-observable. Under what condidtions can such useful categories be created? We propose a solution which requires 1.) that modes or clusters of natural structures are present in the world, and, 2.) that the physical properties of these structures are reflected in the sensory data used by the observer for classification. Given these two constraints, we explore the type of additional knowledge sufficient for the observer to generate an internal representation that makes explicit the natural modes. Finally we develop a formal expression of the object classification problem. |
first_indexed | 2024-09-23T09:27:49Z |
id | mit-1721.1/6443 |
institution | Massachusetts Institute of Technology |
language | en_US |
last_indexed | 2024-09-23T09:27:49Z |
publishDate | 2004 |
record_format | dspace |
spelling | mit-1721.1/64432019-04-11T04:54:44Z Classifying Objects from Visual Information Bobick, Aaron Richards, Whitman Consider a world of 'objects.' Our goal is to place these objects into categories that are useful to the observer using sensory data. One criterion for utility is that the categories allow the observer to infer the object's potential behaviors, which are often non-observable. Under what condidtions can such useful categories be created? We propose a solution which requires 1.) that modes or clusters of natural structures are present in the world, and, 2.) that the physical properties of these structures are reflected in the sensory data used by the observer for classification. Given these two constraints, we explore the type of additional knowledge sufficient for the observer to generate an internal representation that makes explicit the natural modes. Finally we develop a formal expression of the object classification problem. 2004-10-04T14:56:29Z 2004-10-04T14:56:29Z 1986-06-01 AIM-879 http://hdl.handle.net/1721.1/6443 en_US AIM-879 3781558 bytes 2953807 bytes application/postscript application/pdf application/postscript application/pdf |
spellingShingle | Bobick, Aaron Richards, Whitman Classifying Objects from Visual Information |
title | Classifying Objects from Visual Information |
title_full | Classifying Objects from Visual Information |
title_fullStr | Classifying Objects from Visual Information |
title_full_unstemmed | Classifying Objects from Visual Information |
title_short | Classifying Objects from Visual Information |
title_sort | classifying objects from visual information |
url | http://hdl.handle.net/1721.1/6443 |
work_keys_str_mv | AT bobickaaron classifyingobjectsfromvisualinformation AT richardswhitman classifyingobjectsfromvisualinformation |