This project is developed using Flutter. You can launch this app by downloading an apk file or running with IDE(VSCode)!
- Enter https://github.com/RememVR-2024-SolutionChallenge/RememVR-client/releases
- Click the latest release
- Download app-release.apk
- Launch this app on Android devices.
Before you start, make sure you have installed the following on your system:
- Flutter SDK
- VS Code
- Dart and Flutter extensions for VS Code
- Android emulator or physical device
- Flutter Configuration : execute 'flutter doctor' and check configuration.
$ git clone https://github.com/RememVR-2024-SolutionChallenge/RememVR-client.git
$ flutter runWe separated our application into two tracks : Care-giver and Care-recipient Care-givers can make VR scenes and care-recipients can experience VR scenes engaging in reminiscence therapy.
RememVR generates avatars representing the patient's cherished individuals. These avatars, combined with place VR scene, offer the patient an immersive recollection of memories. RememVR assists in safeguarding the continuity of the patient's memories. Caregivers can upload a video featuring the person whom they like to transform into an avatar.
RememVR creates VR experiences encapsulating the places intertwined with a patient's memories. We all live with our unique sense of places, as visiting specific locations can evoke memories from particular times. Places in VR will make them feel as if they are physically present in that space. Care-givers can upload videos that showcase distinctive features of the location.
Caregivers can select one or several VR avatars, and one place VR scene to build a completed VR video. After VR video is generated, care-recipient can find out that videos in his/her VR experience list.
Care-recipients can participate in VR video. Our character ‘Brainy’ helps care recipient to follow the steps for VR experience which is a part of the reminiscence therapy.
After the care-recipient completes the VR experience, badges are awarded as part of gamification. These badges can be viewed not only on the care-recipient's home-page but also on the care-giver's home. There are a total of three badge designs, and one of them is randomly assigned.
- Care-giver uploads images of the desired person's face, full body photos, and background.
- The person and background are reborn as VR avatar and scene models through our AI server.
- Among the created VR models, care-giver selects a model that he wants to make into a VR video, modifies the location information on the VR-viewer web page, and transmits it to the server. After the image generation process, the care-recipient can select and view the video that the VR device wants to see.
- We don't have a separate server that stores VR models itself, but we store VR models that care-giver requested to create locally in the app.
- Since the avatar model includes face and body information corresponding to personal information, if another care-giver logs in, it deletes all stored VR models.
- VR models are stored in the {directoryPath}/sample/{vr-model-id}/model.fbx path and location information is modified on the vr-viewer page with the models selected by the care-giver.
- The storage path of the selected VR models is accessed and edited on the vr-viewer page.
- Metadata of all generated VR images is stored in {directoryPath}/sample/vr-video-id/meta-data.json. The image selected by the care-recipient is viewed on the vr-viewer page.
- The corresponding web page may access the metadata storage path of the selected image and the user can view the image on the vr-viewer page.
