Use augmented reality to transform your room into a museum for interactively viewing your NFT collection. Curator generates custom ARKit experiences from your personal NFTs.
- View your token on the OpenSea website/app.
- Copy the entire URL to your clipboard.
- Paste it into the specified textfield on Curator’s homepage.
- Tap the 3D model you want to customize.
- Aim your iPhone at a flat surface (like a table or the floor) and let the camera stabilize.
Curator is a Dockerized web application that can load an NFT onto a collection of 3D models.
Blender is used to pre-build the 3D models and texture them with a placeholder NFT image. All image textures must be stored in the /var/www/html/models/<model_name>/ directory while building & exporting the model. Any mesh intended to hold the user’s NFT image should be textured with a placeholder image named nft.png in that same directory. The model is exported to an FBX file with the filename <model_name>.fbx. Once the export is complete, the entire <model_name> folder can be copied from /var/www/html/models/ to the models/ directory of your Curator project.
Curator’s web application is built with PHP and is used to collect an OpenSea URL and a desired model from the user. It parses the input to determine the contract address and token ID. A query is made to the OpenSea API to gather the token’s metadata and locate the URL of the NFT’s image. The model name and image URL are passed to curator.sh via shell_exec().
This bash script downloads the NFT’s image and uses it to overwrite nft.png for the selected model. When the FBX file is then converted to a USDZ, it will pull the current contents of nft.png and generate the USDZ file with the user’s NFT image instead of the placeholder image.
The generated USDZ model can then be viewed on any iPhone that supports ARKit.
heroku container:login
heroku labs:enable --app nftcurator runtime-new-layer-extract
heroku container:push web --app nftcurator
heroku container:release web --app nftcurator