Powder is the only app offering a complete turnkey solution to automatically detect the best gaming moments, earn rewards from quests and challenges, and compete with other gamers.

Informazioni relative al progetto

POWDER

ID dell’accordo di sovvenzione: 190156232

DOI

10.3030/190156232

Progetto chiuso

Data della firma CE 25 Ottobre 2022

Data di avvio 1 Agosto 2022

Data di completamento 31 Luglio 2024

Finanziato da

• The European Innovation Council (EIC)

Costo totale

€ 9 747 578,75

Contributo UE

€ 2 500 000,00

Coordinato da

UNIQUE ENTERTAINMENT EXPERIENCE SAS
France

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Other (24)

Models creation and integration for new games

By adding new games to our pipeline we are able to offer richer quests and challenges experiences

Challenge feed recommendation system

Our existing RecSys project is reconverted and improved to support quests and challenges engagement

Wallet supports the 3 major blockchains

Continuous effort of adding multiple blockchains / wallets supported

PC configurations management

PC configurations management (continuous)

Integration of 30 detection models on PC

PC app supports 30 detection models

Integration of 20 detection models on PC

Integration of 20 AI detection models to our PC app

Onboarding discovery

The user experience of quests and challenges is optimised on mobile

Setting up a process for prototype testing

Industrialising our prototype testing

Continuous improvements on PC video recorder CPU & GPU usage

Reduce CPU GPU usage on the users devices

Production launch and nominal operation on PC

Crypto wallet is launched on our PC app at nominal operations

Digital assets marketplace test release

In-app test release

Partners due diligence and selection (Crypto wallet)

We identify audit and select one or multiple crypto wallet partners

Leaderboard launch

A leaderboard where users can see their ranking compared to other participants to quests and challenges is relased

Integration of 15 detection models on PC

PC app supports 15 detection models

Security and implementation testing

Security is critical when handling users assets

Reaching 80% fidelity in new models detection

Reaching high levels of fidelity in the new games added to the model

Improvement on the delays to create an integrate new models for new games

In the perspective of scaling our user base, scaling our approach becomes critical

Optimisation of video resolution

Optimisation of video resolution for external sharing

Redesign of the user profile to highlight rewards PC

Redesign of the user profile to highlight rewards user library

Guaranteeing efficient R&D activities

Ensuring an high level view of all R&D activities all along the project timeline

Redesign of the user profile to highlight rewards Mobile

The user profile is redesigned fully to offer a consistent and enjoyable app experience

Reduce crash rate

reduce the crash rate to 5 to offer a seamless experience

Digital assets marketplace product release

Product release in the public app

Set up of automatic bots to share clips, achievements, etc. with other users

Distribution set up of automatic bots to share clips achievements etc with other users

Documents, reports (1)

PhD research thesis

"PhD research thesis ""Analysis of human behavior from video games using deep learning approaches"" (due in 36 months)"

Demonstrators, pilots, prototypes (2)

Production launch and nominal operation on mobile

Crypto wallet is launched on our mobile app at nominal operations

Quests launch on mobile thanks to the mobile recorder rebuild

Quests individual challenges are released on mobile

Websites, patent fillings, videos etc. (1)

Reach 100k Discord users

Reach 100k Discord users on Powder servers / using the Powder discord bot

Pubblicazioni

Peer reviewed articles (2)

Pattern Recognition Letters

Autori: Liam Schoneveld, Alice Othmani, Hazem Abdelkawy
Pubblicato in: Pattern Recognition Letters, 2021, ISSN 0167-8655
Editore: Elsevier
DOI: 10.1016/j.patrec.2021.03.007

Comparing Learning Methodologies for Self-Supervised Audio-Visual Representation Learning

Autori: H. Terbouche, L. Schoneveld, O. Benson, A. Othmani
Pubblicato in: IEEE Access, Numero Volume 10, pp. 41622-41638, 2022, ISSN 2169-3536
Editore: IEEE Access
DOI: 10.1109/ACCESS.2022.3164745

Conference proceedings (2)

Towards a General Deep Feature Extractor for Facial Expression Recognition

Autori: L. Schoneveld, A. Othmani
Pubblicato in: 2021 IEEE International Conference on Image Processing (ICIP), Numero 2339-2342, 2021, ISSN 2381-8549
Editore: IEEE Computer Society
DOI: 10.1109/ICIP42928.2021.9506025

Multi-Annotation Attention Model for Video Summarization

Autori: Hacene Terbouche, Maryan Morel, Mariano Rodriguez, Alice Othmani
Pubblicato in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2023, ISSN 3142-3151
Editore: IEEE/CVF (Computer Vision Foundation)