Valossa’s unparalleled visual recognition capabilities also work with images.

OULU, FINLAND – Using the same visual capabilities as the Video Recognition API, Valossa rolls out a new demo for image recognition.

Recent developments on Valossa AI for video moderation has triggered demand to bring the powerful recognition capabilities also for still media, i.e. images. Valossa image recognition technology is built for the purpose of making it easier to make pictorial content safer, more appropriate and better described for online media consumption.

We have built a demo application that lets anyone to try out Valossa visual recognition engines for their content with easy upload and inspection tools. You can see how Valossa AI extracts intelligence from image data for a broad range of content exposing the objects, styles, scenes, people, activities and even color composition for style matching. Comprehensive visual inspection of sexual, sensual, explicit, violent, threatening, and weapon entities helps automating content moderation processes and profiling content assets with high volumes of images. Together with contextual tags of objects and scene activities, Valossa AI is able to separate between use of knives as a tool in cooking or in violent situations.

As with the Video Recognition API, Valossa AI recognizes a huge variety of visual concepts. For images, these concepts include:

  • Appearing objects and image attributes
  • Faces and gender
  • Locations
  • Potentially inappropriate or adult-oriented content
  • Dominant color structure

…and more.

The demo is hooked with our pre-trained celebrity face gallery, which recognizes over a thousand globally trending celebrities.

We invite you to try the demo yourself. Follow the link below and upload any .jpg or .png and see the results immediately.

IMAGE RECOGNITION API DEMO

Contact
Lauri Penttilä
Marketing & Communications Manager
lauri.penttila@valossa.com