Related Experiment Video
Updated: May 29, 2026
08:25
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
Semantic description of aerial images using stochastic labeling
1MEMBER, IEEE, Image Processing Institute, University of Southern California, Los Angeles, CA 90007; Institut National de Recherche en Informatique et en Automatique, Le Chesnay, Fr.
IEEE Transactions on Pattern Analysis and Machine Intelligence
|August 27, 2011
Summary
This study applies stochastic labeling to symbolic image description, introducing a novel method for computing likelihoods and compatibilities. The technique minimizes a global criterion for improved image matching and description accuracy.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Pattern Recognition
Background:
- Symbolic image description is crucial for understanding visual data.
- Existing symbolic matching procedures require modification for advanced labeling techniques.
Purpose of the Study:
- To apply stochastic labeling to general symbolic image description.
- To develop and describe a method for computing initial likelihoods and compatibilities.
- To introduce a labeling procedure that minimizes a global criterion iteratively.
Main Methods:
- A modified symbolic matching procedure to derive data for stochastic labeling.
- Iterative minimization of a global criterion within the labeling process.
- Comparison of the proposed stochastic labeling technique with existing matching methods.
Main Results:
- The developed method successfully computes initial likelihoods and compatibilities.
- The iterative global criterion minimization enhances the labeling procedure.
- Performance evaluation on two distinct scenes demonstrates the technique's efficacy.
Conclusions:
- Stochastic labeling offers a robust approach to symbolic image description.
- The modified method provides a significant improvement over simpler labeling techniques.
- The presented results validate the applicability and effectiveness of the proposed approach.
Related Concept Videos
01:29Application of Linearization and Approximation
A drone flying through complex terrain often relies on more than one sensing method to estimate small changes in altitude. Along with direct measurements, air pressure provides a useful indirect indicator of vertical movement. Atmospheric pressure decreases as altitude increases, and this relationship is commonly described using an exponential model. Although accurate, converting pressure measurements into altitude values requires calculations that are too complex to perform repeatedly during...
02:16Light Acquisition
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.