Compound Effects of Top-Down and Bottom-up Influences on Visual Attention During Action Recognition
Abstract
The limited visual and computational resources available during the perception of a human action makes a visual attention mechanism essential.In this paper we propose an attention mechanism that combines the saliency of top-down (or goaldirected) elements, based on multiple hypotheses about the demonstrated action, with the saliency of bottom-up (or stimulus-driven) components.Furthermore, we use the bottom-up part to initialise the top-down, hence resulting in a selection of the behaviours that rightly require the limited computational resources.This attention mechanism is then combined with an action understanding model and implemented on a robot, where we examine its performance during the observation of object-directed human actions.
Cite
Text
Khadhouri and Demiris. "Compound Effects of Top-Down and Bottom-up Influences on Visual Attention During Action Recognition." International Joint Conference on Artificial Intelligence, 2005.Markdown
[Khadhouri and Demiris. "Compound Effects of Top-Down and Bottom-up Influences on Visual Attention During Action Recognition." International Joint Conference on Artificial Intelligence, 2005.](https://mlanthology.org/ijcai/2005/khadhouri2005ijcai-compound/)BibTeX
@inproceedings{khadhouri2005ijcai-compound,
title = {{Compound Effects of Top-Down and Bottom-up Influences on Visual Attention During Action Recognition}},
author = {Khadhouri, Bassam and Demiris, Yiannis},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2005},
pages = {1458-1463},
url = {https://mlanthology.org/ijcai/2005/khadhouri2005ijcai-compound/}
}