The improvement and automation of the ‘VirtuAI Annotation’ system — adapting the tools to the market requirements for AI technologies (Q2732339): Difference between revisions
Jump to navigation
Jump to search
(Changed label, description and/or aliases in en: translated_label) |
(Created claim: intervention field (P888): Research and innovation processes in SMEs (including voucher schemes, process, design, service and social innovation) (Q200704), adding category of intervention) |
||
| Property / intervention field | |||
| Property / intervention field: Research and innovation processes in SMEs (including voucher schemes, process, design, service and social innovation) / rank | |||
Normal rank | |||
Revision as of 12:00, 8 June 2021
Project Q2732339 in Croatia
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | The improvement and automation of the ‘VirtuAI Annotation’ system — adapting the tools to the market requirements for AI technologies |
Project Q2732339 in Croatia |
Statements
1,342,847.52 Croatian kuna
0 references
1,598,628.0 Croatian kuna
0 references
85.0 percent
0 references
25 October 2019
0 references
7 April 2021
0 references
PROTOSTAR LABS društvo s ograničenom odgovornošću za proizvodnju i usluge
0 references
45°41'4.20"N, 18°24'8.50"E
0 references
31551
0 references
Umjetna inteligencija zasniva se na procesu učenja pri kojem inteligentni uređaji, uz pomoć računalnih algoritama i velikog broja slika iz stvarnog okruženja, uče raspoznavati objekte u stvarnom okruženju. Za potrebe procesa učenja nužno je osigurati veliki broj precizno označenih slika, što je novim tvrtkama koje se bave ovom tematikom prevelik financijski izdatak, a sam proces označavanja slika je dugotrajan, skup i podložan pogreškama uslijed ljudskog faktora. Provedbom projekta, omogućit će se stvaranje virtualnih podatkovnih skupova slika koji će se koristiti u procesu strojnog učenja, što će značajno smanjiti cijenu i vrijeme potrebno za njihovo kreiranje, čime će se osigurati pozicija na tržištu umjetne inteligencije. (Croatian)
0 references
AI is based on a learning process where intelligent devices, using computer algorithms and large numbers of images from a real environment, learn to distinguish their facilities in a real-life environment. For the purposes of the learning process, it is necessary to provide for a large number of accurately labelled images, which are new firms dealing with the issue of excessive financial expenditure, and the process to identify images is long-lasting, costly and prone to error as a result of the human factor. The implementation of the project will allow the creation of virtual data sets of images to be used in the machine learning process, which will significantly reduce the price and the time needed to create them, thus securing the position in the AI market. (English)
3 June 2021
0 references
Identifiers
KK.03.2.2.04.0328
0 references