Modeling the influence of task on attention. Navalpakkam, V. & Itti, L.
Paper abstract bibtex We propose a computational model for the task-specific guidance of visual attention in real-world scenes. Our model emphasizes four aspects that are important in biological vision: determining task-relevance of an entity, biasing attention for the low-level visual features of desired targets, recognizing these targets using the same low-level features, and incrementally building a visual map of task-relevance at every scene location. Given a task definition in the form of keywords, the model first determines and stores the task-relevant entities in working memory, using prior knowledge stored in long-term memory. It attempts to detect the most relevant entity by biasing its visual attention system with the entityÕs learned low-level features. It attends to the most salient location in the scene, and attempts to recognize the attended object through hierarchical matching against object representations stored in long-term memory. It updates its working memory with the task-relevance of the recognized entity and updates a topographic task-relevance map with the location and relevance of the recognized entity. The model is tested on three types of tasks: single-target detection in 343 natural and synthetic images, where biasing for the target accelerates target detection over twofold on average; sequential multiple-target detection in 28 natural images, where biasing, recognition, working memory and long term memory contribute to rapidly finding all targets; and learning a map of likely locations of cars from a video clip filmed while driving on a highway. The modelÕs performance on search for single features and feature conjunctions is consistent with existing psychophysical data. These results of our biologically-motivated architecture suggest that the model may provide a reasonable approximation to many brain processes involved in complex task-driven visual behaviors.
@article{
title = {Modeling the influence of task on attention},
type = {article},
identifiers = {[object Object]},
keywords = {Attention,Bottom-up,Object detection,Recognition,Scene analysis,Task-relevance,Top-down},
id = {cb89dc40-bd68-3f0d-9c96-7c9824db361b},
created = {2017-09-01T15:54:06.416Z},
file_attached = {true},
profile_id = {80da7853-f7b7-36a9-8e4c-d7ddb2d9e538},
group_id = {a2333ea3-15a4-3d40-8d36-f0d9590ca926},
last_modified = {2017-09-01T15:54:06.531Z},
read = {false},
starred = {false},
authored = {false},
confirmed = {false},
hidden = {false},
abstract = {We propose a computational model for the task-specific guidance of visual attention in real-world scenes. Our model emphasizes four aspects that are important in biological vision: determining task-relevance of an entity, biasing attention for the low-level visual features of desired targets, recognizing these targets using the same low-level features, and incrementally building a visual map of task-relevance at every scene location. Given a task definition in the form of keywords, the model first determines and stores the task-relevant entities in working memory, using prior knowledge stored in long-term memory. It attempts to detect the most relevant entity by biasing its visual attention system with the entityÕs learned low-level features. It attends to the most salient location in the scene, and attempts to recognize the attended object through hierarchical matching against object representations stored in long-term memory. It updates its working memory with the task-relevance of the recognized entity and updates a topographic task-relevance map with the location and relevance of the recognized entity. The model is tested on three types of tasks: single-target detection in 343 natural and synthetic images, where biasing for the target accelerates target detection over twofold on average; sequential multiple-target detection in 28 natural images, where biasing, recognition, working memory and long term memory contribute to rapidly finding all targets; and learning a map of likely locations of cars from a video clip filmed while driving on a highway. The modelÕs performance on search for single features and feature conjunctions is consistent with existing psychophysical data. These results of our biologically-motivated architecture suggest that the model may provide a reasonable approximation to many brain processes involved in complex task-driven visual behaviors.},
bibtype = {article},
author = {Navalpakkam, Vidhya and Itti, Laurent}
}
Downloads: 0
{"_id":"ySPan5zRHZBFYuEbJ","bibbaseid":"navalpakkam-itti-modelingtheinfluenceoftaskonattention","downloads":0,"creationDate":"2017-09-01T15:59:40.298Z","title":"Modeling the influence of task on attention","author_short":["Navalpakkam, V.","Itti, L."],"year":null,"bibtype":"article","biburl":null,"bibdata":{"title":"Modeling the influence of task on attention","type":"article","identifiers":"[object Object]","keywords":"Attention,Bottom-up,Object detection,Recognition,Scene analysis,Task-relevance,Top-down","id":"cb89dc40-bd68-3f0d-9c96-7c9824db361b","created":"2017-09-01T15:54:06.416Z","file_attached":"true","profile_id":"80da7853-f7b7-36a9-8e4c-d7ddb2d9e538","group_id":"a2333ea3-15a4-3d40-8d36-f0d9590ca926","last_modified":"2017-09-01T15:54:06.531Z","read":false,"starred":false,"authored":false,"confirmed":false,"hidden":false,"abstract":"We propose a computational model for the task-specific guidance of visual attention in real-world scenes. Our model emphasizes four aspects that are important in biological vision: determining task-relevance of an entity, biasing attention for the low-level visual features of desired targets, recognizing these targets using the same low-level features, and incrementally building a visual map of task-relevance at every scene location. Given a task definition in the form of keywords, the model first determines and stores the task-relevant entities in working memory, using prior knowledge stored in long-term memory. It attempts to detect the most relevant entity by biasing its visual attention system with the entityÕs learned low-level features. It attends to the most salient location in the scene, and attempts to recognize the attended object through hierarchical matching against object representations stored in long-term memory. It updates its working memory with the task-relevance of the recognized entity and updates a topographic task-relevance map with the location and relevance of the recognized entity. The model is tested on three types of tasks: single-target detection in 343 natural and synthetic images, where biasing for the target accelerates target detection over twofold on average; sequential multiple-target detection in 28 natural images, where biasing, recognition, working memory and long term memory contribute to rapidly finding all targets; and learning a map of likely locations of cars from a video clip filmed while driving on a highway. The modelÕs performance on search for single features and feature conjunctions is consistent with existing psychophysical data. These results of our biologically-motivated architecture suggest that the model may provide a reasonable approximation to many brain processes involved in complex task-driven visual behaviors.","bibtype":"article","author":"Navalpakkam, Vidhya and Itti, Laurent","bibtex":"@article{\n title = {Modeling the influence of task on attention},\n type = {article},\n identifiers = {[object Object]},\n keywords = {Attention,Bottom-up,Object detection,Recognition,Scene analysis,Task-relevance,Top-down},\n id = {cb89dc40-bd68-3f0d-9c96-7c9824db361b},\n created = {2017-09-01T15:54:06.416Z},\n file_attached = {true},\n profile_id = {80da7853-f7b7-36a9-8e4c-d7ddb2d9e538},\n group_id = {a2333ea3-15a4-3d40-8d36-f0d9590ca926},\n last_modified = {2017-09-01T15:54:06.531Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {false},\n hidden = {false},\n abstract = {We propose a computational model for the task-specific guidance of visual attention in real-world scenes. Our model emphasizes four aspects that are important in biological vision: determining task-relevance of an entity, biasing attention for the low-level visual features of desired targets, recognizing these targets using the same low-level features, and incrementally building a visual map of task-relevance at every scene location. Given a task definition in the form of keywords, the model first determines and stores the task-relevant entities in working memory, using prior knowledge stored in long-term memory. It attempts to detect the most relevant entity by biasing its visual attention system with the entityÕs learned low-level features. It attends to the most salient location in the scene, and attempts to recognize the attended object through hierarchical matching against object representations stored in long-term memory. It updates its working memory with the task-relevance of the recognized entity and updates a topographic task-relevance map with the location and relevance of the recognized entity. The model is tested on three types of tasks: single-target detection in 343 natural and synthetic images, where biasing for the target accelerates target detection over twofold on average; sequential multiple-target detection in 28 natural images, where biasing, recognition, working memory and long term memory contribute to rapidly finding all targets; and learning a map of likely locations of cars from a video clip filmed while driving on a highway. The modelÕs performance on search for single features and feature conjunctions is consistent with existing psychophysical data. These results of our biologically-motivated architecture suggest that the model may provide a reasonable approximation to many brain processes involved in complex task-driven visual behaviors.},\n bibtype = {article},\n author = {Navalpakkam, Vidhya and Itti, Laurent}\n}","author_short":["Navalpakkam, V.","Itti, L."],"urls":{"Paper":"http://bibbase.org/service/mendeley/80da7853-f7b7-36a9-8e4c-d7ddb2d9e538/file/13982c22-812f-d536-732e-8d3817a55ae4/Modeling_the_influence_of_task_on_attention.pdf.pdf"},"bibbaseid":"navalpakkam-itti-modelingtheinfluenceoftaskonattention","role":"author","keyword":["Attention","Bottom-up","Object detection","Recognition","Scene analysis","Task-relevance","Top-down"],"downloads":0},"search_terms":["modeling","influence","task","attention","navalpakkam","itti"],"keywords":["attention","bottom-up","object detection","recognition","scene analysis","task-relevance","top-down"],"authorIDs":[]}