Inimteadmiste kasutamine, loomaks sarnasushinnangute andmebaasi tehisnärvivõrkude arendamiseks
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Viimaste aastate jooksul on kasvanud huvi selle vastu, kuidas inimesed tajuvad ja kategoriseerivad pilte. Loodetakse, et seda informatsiooni saab kasutada selleks, et arendada ja täiustada tehislikke pildiklassifitseerimissüsteeme. Üks suund, kuidas mõista inimteadmist piltide klassifitseerimisel, on uurida erinevate objektide sarnasusi. Kuid suure hulga võimalikke objektide tõttu, mis tekitaksid veel suurema hulga võimalikke võrdlusi, on klassikaliste eksperimentide võimalused sarnasusruumide konstrueerimiseks piiratud.
Antud projekti eesmärgiks oli luua veebirakendus, millele pääseks ligi suur hulk inimesi üle maailma, kes saaksid panustada antud projekti sihti. Veebirakendus võimaldab kasutajatel hinnata loomulikke pilte tajutava sarnasuse põhjal. See toodaks suure ja usaldusväärse andmehulga piltide sarnasustest, millele pääseksid ligi ja mida saaksid kasutada nii teadlased kui ka pildiklassifitseerimise eksperdid üle maailma. Kogutud andmete põhjal koostatakse sarnasusmaatriks. Antud rakendus ning sellega kaasnev andmebaas aitavad meil loodetavasti paremini mõista inimnägemist. Samuti võib antud andmehulk kasuks tulla objektide kategoriseerimiseks loodud tehisintellekti algoritmide arendamisel.
In recent years, there has been a growing interest in understanding the way in which humans encode and categorise visual images. One hope is that this knowledge can be translated into improvements in artificial agents that carry out image classification automatically. One approach for closing in on the code of human object categorisation is to understand the relational coding of different objects in terms of their similarity. However, due to the large number of possible objects and their even larger number of possible comparisons, classical experimental approaches are limited in building representational similarity spaces. The goal of this project was to create a web-application, which could be accessed by a large number of people around the world, who could contribute to the goal of the project. The webpage allows subjects to rate natural images according to their perceived similarity. This yields a large and reliable dataset of image ratings that could be accessed and used by experimenters, but also by image classification experts worldwide. Based on the dataset a similarity matrix can be constructed. The webpage and the corresponding database will hopefully advance our understanding of human vision and may one day be used to improve artificial intelligence algorithms designed to categorise objects.
In recent years, there has been a growing interest in understanding the way in which humans encode and categorise visual images. One hope is that this knowledge can be translated into improvements in artificial agents that carry out image classification automatically. One approach for closing in on the code of human object categorisation is to understand the relational coding of different objects in terms of their similarity. However, due to the large number of possible objects and their even larger number of possible comparisons, classical experimental approaches are limited in building representational similarity spaces. The goal of this project was to create a web-application, which could be accessed by a large number of people around the world, who could contribute to the goal of the project. The webpage allows subjects to rate natural images according to their perceived similarity. This yields a large and reliable dataset of image ratings that could be accessed and used by experimenters, but also by image classification experts worldwide. Based on the dataset a similarity matrix can be constructed. The webpage and the corresponding database will hopefully advance our understanding of human vision and may one day be used to improve artificial intelligence algorithms designed to categorise objects.