Web3 Dev Tools

Web3 Dev Tools consists of dev tools powered by Web3 technologies. As of now, it includes Container Registry & NFT Detection using machine learning models (AI Hub).

Container Registry allows you to push and pull docker images from IPFS and Filecoin. It provides a native docker integration via a custom docker registry server (v2). It is powered by web3.storage and OrbitDB. The docker image layers and the manifest file are pushed and pulled from IPFS and Filecoin using web3.storage. The OrbitDB is used to maintain the mapping of IPFS hash for the docker image layers and the manifest file.

NFT Detection uses Tensorflow Keras to generate a machine learning model to detect similar NFTs based on the uploaded image. The Graph protocol is used to generate the dataset required for training the machine learning model. The generated model is then pushed to IPFS / Filecoin using web3.storage. Once the model server is started, then the model is pulled from IPFS and used for the inference. NFT Detection model can be used using a UI or using a REST API. As of now, it only supports the CryptoPunks NFT. The dataset can be updated to handle other NFTs too.

Use cases

• Search NFT using an image
• Similar NFT Detection
• Detect smart contract address and token id for an NFT using an image
• Counterfeit NFT detection
• Detect NFT owner address based on twitter profile image (If CryptoPunks NFT is used as a profile image)

Web3 Dev Tools consists of dev tools powered by Web3 technologies. As of now, it includes Container Registry & NFT Detection using machine learning models (AI Hub).

Container Registry allows you to push and pull docker images from IPFS and Filecoin. It provides a native docker integration via a custom docker registry server (v2). It is powered by web3.storage and OrbitDB. The docker image layers and the manifest file are pushed and pulled from IPFS and Filecoin using web3.storage. The OrbitDB is used to maintain the mapping of IPFS hash for the docker image layers and the manifest file.

NFT Detection uses Tensorflow Keras to generate a machine learning model to detect similar NFTs based on the uploaded image. The Graph protocol is used to generate the dataset required for training the machine learning model. The generated model is then pushed to IPFS / Filecoin using web3.storage. Once the model server is started, then the model is pulled from IPFS and used for the inference. NFT Detection model can be used using a UI or using a REST API. As of now, it only supports the CryptoPunks NFT. The dataset can be updated to handle other NFTs too.

Use cases

• Search NFT using an image

• Similar NFT Detection

• Detect smart contract address and token id for an NFT using an image

• Counterfeit NFT detection

• Detect NFT owner address based on twitter profile image (If CryptoPunks NFT is used as a profile image)