UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Remi: An AI Conversational Agent to Reminisce with Photos

Bhatnagar, Tigmanshu; Patel, Dilisha; (2023) Remi: An AI Conversational Agent to Reminisce with Photos. Presented at: Human-Computer Interaction – INTERACT 2023 19th IFIP TC13 International Conference, York, UK. Green open access

[thumbnail of AITOA23_paper_2.pdf]
Preview
Text
AITOA23_paper_2.pdf - Published Version

Download (9MB) | Preview

Abstract

We propose a conversational AI agent design concept called Remi to signify and celebrate reminiscence. Remi would be a networked digital photo frame. It can engage with older adults and simulate interactions with their loved ones through recognition of uploaded photos, conversation generation around the photographs and interactivity through natural language processing. Remi will allow users to upload old and recent photographs and videos into the device and even voice record anecdotes and memories around the imagery. Old photographs could be restored and turned into a few seconds of ‘live photos’ based on conversations with the user to reminisce with them. Beyond this processing, Remi will have a built-in personality to simulate and prompt conversations around a photograph. This would mean that older adults can interact and talk to Remi whenever they want and converse on demand.

Type: Conference item (Presentation)
Title: Remi: An AI Conversational Agent to Reminisce with Photos
Event: Human-Computer Interaction – INTERACT 2023 19th IFIP TC13 International Conference
Location: York, UK
Dates: 28 August - 01 September 2023
Open access status: An open access version is available from UCL Discovery
Publisher version: https://interact2023.org/
Language: English
Keywords: AI, Conversational, Old People
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science
URI: https://discovery.ucl.ac.uk/id/eprint/10178480
Downloads since deposit
30Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item