A new visual medium

to extract, search & share visual data.
Powered by Artificial Intelligence

Ease of Screen Recording


Capture tab, window or entire screen.


Simultaneously record tab audio.


Runs as a website in Chrome & Edge.

Power of Artificial Intelligence


We enable both text & visual search across highlights, clips and entire audio / video streams. We extract objects such as (cars, faces, etc.) and make them searchable. Additionally you can incorporate metadata collected directly from captured tab and create annotations.


Build automations that leverage features and data extracted by built-in models. Customize SuperVisual to suit needs of a particular workflow or website being watched. Export recorded sessions to a file to share and collaborate with others.


We leverage latest features in Chrome & Edge browsers, to deliver a private and secure experience. All recordings are stored inside the browser using IndexedDB. ML models run locally with TensorflowJS using WASM & WebGL. We do not require any extension to be installed.



No Sign-up required
  • ML models run locally
  • Share & restore with file export
  • Local image and text search
  • Data is stored within browser
Open Watcher


Work with your team
Additional features
  • ML models on cloud GPUs/TPUs
  • Live collaboration & sharing  
  • Full text & vector search engine
  • Private Google Cloud project

Community version

  • ML models on cloud GPUs/TPUs
  • Private Google Cloud project

Leverage power of Artificial Intelligence