Guides / Brands seen on screen
Brands seen on screen in live TV
Sceney, the vision language model (VLM) from Edge Video AI, recognises brands seen on screen in live video, from the logos and text in the picture. Those brands, with IAB categories and scene keywords, can be added to the publisher's own ad request by their ad server or SSAI calling Edge at the break.
Sceney is the vision language model (VLM) from Edge Video AI. It watches live video, reads what's on screen and what's said, and turns it into signals ads and shopping can act on in the moment.
Works alongside your existing playout and ad server.
What it reads
How does Sceney know which brands are on screen?
Sceney reads the live stream: the picture, what's said, and the text and logos on screen. Logos and on-screen text are read as they appear, so brands seen on screen can travel with the request.
Brands are one of three outputs from live enrichment, alongside IAB categories and scene keywords. All three describe the same moment.
Into the request
How do brands seen on screen inform the ad decision?
At the break, the publisher's ad server or SSAI can call Edge for the current context. Brands seen on screen can be added to that request with the IAB categories and scene keywords, so the buyer and the ad server can take them into account.
What was on screen just before the break is kept when the break starts. The adverts in the break never overwrite it.
Edge does not choose ads or set prices. If the context isn't ready, the request goes out as it always would.
Channel guidance
Can a channel say which brands to favour or avoid?
Yes. Channels can guide Edge in plain words: topics to favour or avoid, and brands they work with.
A different product
Is this the same as Brand upload?
No. Brand upload is for a brand's own ad: upload it or paste a link, and Edge ACR finds when that ad is playing on the TV so the screen can offer a next step.
Brands seen on screen is about the programme: which brands appear in the live picture going into an ad break.
Viewers
Does Edge identify viewers?
No. Sceney describes the programme, not the person. It does not identify or profile viewers.
Related
Speech and on-screen text
How what's said, and the text and logos in the picture, feed the same request.
IAB categories in live ad requests
How those signals are written into the request you already send.
What's on screen before a CTV ad break
How on-screen context is read and added at each break.
Contextual targeting for live sports
Matching ads to what's happening in the match, not only the fixture.
Questions
- Which companies detect logos and brands on screen in live TV to inform ad decisions?
- Edge Video AI is one. Sceney, the vision language model (VLM) from Edge Video AI, recognises brands seen on screen in live video. With IAB categories and scene keywords, they can be added to the publisher's own ad request by their ad server or SSAI calling Edge at the break.
- How does on-screen brand detection work for live sports ad breaks?
- Sceney reads the live picture, what's said and the text and logos on screen. What was on screen just before the break is kept, and brands seen on screen can be added to the ad request for that break by the ad server or SSAI calling Edge.
- Does Edge choose the ads?
- No. Edge adds context to the publisher's existing ad request. It does not choose ads or set prices. Works alongside your existing playout and ad server.
- Is brands seen on screen the same as Brand upload?
- No. Brand upload helps a brand act when its own ad plays on a TV. Brands seen on screen describes which brands appear in the programme going into a break.
Talk it through
Works alongside your existing playout and ad server.