WatchDog, IP Protection Service in Web3


Intellectual property protection service in Web3

  • Blockchain
  • Web
  • Machine Learning
  • AI
  • NFT
  • Web3

About the client

CheckNFT is a fully-fledged NFT ecosystem comprising a set of high-end products and services aimed at making the entire NFT landscape safer and helping users enjoy fruitful cooperation in the Web3 space.


PixelPlex has delivered an AI-powered IP protection platform that is set to help NFT creators, projects, brands, and marketplaces to track and prevent the infringement of their intellectual property in the Web3 space.


  • Data scientists2
  • Data engineer
  • Data analyst
  • Full-stack developers2
  • ML engineer
  • Front-end developer
  • DevSecOps

Business opportunity

Non-fungible tokens have been growing in popularity, but this also means they have become a tempting target for savvy fraudsters. As a result, NFT creators and projects as well as brands and marketplaces have had to deal with a range of violations of their intellectual properties, including:

  • Attention iconAttacks via compromised social media accounts
  • Attention iconTrademark infringement
  • Attention iconUnauthorized name/art usage
  • Attention iconPlagiarized work

As these IP violations lead to drastic financial losses and prevent the wider adoption of NFTs, a sophisticated and robust solution is needed to help confront the challenge.

Tweets of notable NFT-related accounts about the frauds

Project goals

  • 1

    Develop a high-end solution that would allow NFT creators, projects, etc. to spot and prevent the infringement of their intellectual property

  • 2

    Incorporate the ability to identify IP violations by image, trademark, project, brand, or creator name

  • 3

    Enable users to generate comprehensive DMCA reports on authorized IP usage directly on the platform

  • 4

    Provide an understandable and easy-to-navigate user interface

  • 5

    Interconnect the platform with other CheckNFT ecosystem services and apps to allow for exceptional UX

Work done

  • Cutting-edge web platform helping users to identify and prevent the infringement of their intellectual property in Web3

  • Deployment and integration of advanced computer vision and NLP models for detecting potential fakes and duplicates

  • Built-in data layer containing the most complete data about collections, NFTs, events, and wallets

  • Intuitive user interface empowering users to keep track of their intellectual property in Web3

Shapes of three NFT items from the Bored Ape Yacht Club collection


The PixelPlex team has developed a specialized service platform for NFT creators, brands, and marketplaces that seeks to help detect and impede duplicates, copycats, and trademark infringements. It achieves this by leveraging the powers of computer vision and NLP AI models. The solution makes it possible to prevent reputational and financial losses and to safeguard NFT communities.

WatchDog takes >1 minute to capture a new event.

  • Rocket iconEvents parsed
  • Lightning iconNFT collections in total
  • Fire iconNFT assets processed on Ethereum
Shapes of three NFT items from the Bored Ape Yacht Club collection

Keep your web3 assets safe — request WatchDog demo now

Main capabilities of WatchDog

With WatchDog's advanced capabilities, users can seamlessly differentiate between authentic assets and fake logos, detect wash trading anomalies, and spot counterfeit NFTs, preserving the integrity of their web3 ecosystems.

Fake logo detection

WatchDog safeguards against unauthorized logo use with an extensive logo database and advanced detection algorithms.

It scans and analyzes an image to detect any logo patterns within it. Using Contrastive Language-Image Pre-Training (CLIP) for image-text embedding, it then identifies and classifies logos. Our customized CLIP model, trained on combined datasets from YOLO, Frickl, and LogoDet-3K, ensures optimal performance.

We employ the Faiss library for efficient logo comparison, offering benefits like fast similarity searches and large vector clustering, making it a crucial asset in our work.

WatchDog detects fake logos of famous brands

Wash trading detection

WatchDog offers comprehensive token transaction analysis, identifying potential wash trading or scams through three key steps:

  • Data preprocessing. Eliminating duplicates and marking transactions and accounts with advanced data cleaning algorithms to ensure accuracy of the subsequent analysis.
  • Graph creation. Constructing directed graphs of account interactions using advanced graph theory and network analysis, reducing false positives.
  • Transaction clustering. Clustering transaction sequences based on potential motives, labeling each cluster with a motive step, unique ID, risk assessment weight, and associated accounts.
WatchDog detects wash trading cases after a thorough transaction analysis

Fake NFTs detection

Our tool distinguishes real NFTs from fakes by utilizing Convolutional Neural Networks (CNNs) to extract high-dimensional image features.

These features are then converted into embeddings using transfer learning models, enabling WatchDog to capture essential image characteristics for efficient comparison.

This abstract comparison allows it to detect similarities that would be missed in a pixel-to-pixel analysis, thus enabling the detection of sophisticated counterfeits with subtle alterations.

WatchDog detects fake NFTs after a thorough analysis

Steps to detecting and combating IP infringements

IP tracking by different parameters

The platform enables users to specify the parameters by which they want to track and protect their IP. These can be a project, brand, artist, or creator name. At the same time, users have the ability to upload an image of their artwork or search for it by collection name or contract address.
The search bar interface of WatchDog

AI-enabled fake and duplicates monitoring

WatchDog monitors blockchain in real-time, analyzes all existing NFTs, and compares them to the IP that the user has indicated. Thanks to the integrated AI engine, the solution quickly spots potential unauthorized IP usage and provides comprehensive information that helps prioritize the next steps and protect the user’s work and community.
The search bar interface of WatchDog

Instant alerts about potential IP infringement

If WatchDog detects copycats or name exploitation, the platform immediately notifies a user via a preferred channel, such as Discord, Telegram, Twitter, email, or text messages. Importantly, it allows a user to set up notifications to be sent directly to the community or to the user’s team for pre-moderation.
The search bar interface of WatchDog

Complete reports on unauthorized IP usage

Once WatchDog has spotted a fake, it gives the user the option of generating a DMCA report which will consist of a number of helpful details necessary to prevent unauthorized usage of their work and remove fakes on marketplaces and social networks, either ad hoc or automatically.
The search bar interface of WatchDog
The search bar interface of WatchDog

How the solution leverages computer vision

Our experts took care to implement sophisticated computer vision algorithms that would enable WatchDog to effectively monitor blockchain for potential IP infringement. The integrated models are based on the autoencoder architecture with the four-step education stage:

  • 01

    Autoencoder training with triplet loss on non-clustered data

  • 02

    Preparation of a new dataset of clustered data using the model from stage one

  • 03

    Training of a new "strong" autoencoder which is predicated on the triplet data where the anchor, positive and negative samples are all from the same cluster

  • 04

    Training of the model needed to obtain embeddings of a whole collection of NFT images. The model has an autoencoder architecture, which was trained by minimizing the distance between the embedding of a random image from a collection and the average embedding of all images in this collection

As a result, our engineers worked out two models:

  • Strong autoencoder, which was trained on adjacent images

  • Embedding model of a whole NFT collection — a model that is trained by minimizing the distance between the embedding of one image from the collection and the average embedding of the entire collection

Project features

An icon of a user profile

Convenient search of user’s IP

An icon of Ethereum

Support for Ethereum assets (Solana coming soon)

An icon of three layers

Built-in data layer with data for over 82 million NFTs, events, etc.

An icon of an eye

Computer vision and NLP models monitoring blockchain in real time

An icon of an electric battery

Ability to create DMCA reports

An icon of a magnifier

Accurate infringement detection

An icon of an infographic

First-hand, real-time, fork-tolerant data

An icon of a shield

Guaranteed transparent cooperation

Technologies used