When dealing with huge amounts of creatives (video and images) it might be hard to identify which creative approaches work the best since actual performance is usually evaluated on a single creative.
Filonov allows you to answer the question - what are the similarities between my media (being image or videos). Simply links to your media, specify tagger and get results in no time!
Project Filonov consists of two major parts:
- Data Pipeline
- Data Visualization
Data Pipeline is currently a set of command-line tools which target is creating a JSON file with graph of assets (images/video) grouped in clusters based on similarity. Additionally asset nodes enriched with different metrics from Google Ads API and YouTube API to describe features and performance metrics of creatives.
Data Visualization is a web application uniformly accessible for all users from http://filonov-ai.web.app. It's deployed to Firebase Static Hosting and implemented as serverless web application where users can open data files generated with the Data Pipeline.
Entry point of the Data Pipeline is a CLI tool called filonov
that performs necessary steps to generate input for Filonov UI.
Check the documentation on how to use filonov.
To start using the visualization application we need to prepare graph data using the Filonov Data Pipeline first. As a result we must get a json file, which you'll open locally or via a remote link.
The web application is available at: http://filonov-ai.web.app. It's completely serverless, as such there's no any authorization for accessing it, it's publicly available.
This is not an officially supported Google product. This project is not eligible for the Google Open Source Software Vulnerability Rewards Program.