With the rise of short-form video content, livestream video is also increasingly rising through CTV and social media channels including Facebook, Instagram, Periscope and LinkedIn.
Live broadcast is particularly dangerous for brands, and yet the opportunity to reach a highly engaged audience warrants a consideration – assuming brand safety risks are mitigated. Without advanced AI-based tools and automation, there is no other practical way to intervene if a brand-safety risk emerges: it’s already far too late. Netra has introduced a solution that can process live stream content and immediately return both the context and brand safety risks systematically.
Why our API is game-changing
- Ability to reduce risk to brands in live broadcast content
- A single endpoint to classify streaming video content for both buy-side and sell-side ecosystem partners to enable a consistent taxonomy, improve monetization opportunities, and improve targeting of the right audiences with relevant content.
- Delivers granular details about scenes that cannot be analyzed efficiently without AI-based technology.
How it’s implemented
- We recognize both objects and text at the frame level, which is then used to construct the “aboutness” of a scene.
- Timestamps are provided for scene level detail, with the associated metadata.
- Outputs are expressed in a multi-label classification format which include people, places, objects, activities, emotions, and affinities.
- Support for industry standards like IAB and GARM are provided
- The output is a file, which is human readable and constructed in a manner that is consistent and easily parsed for your use cases.
- In terms of interpreting results, we provide all applicable categories and segments with a score. This score is a strength of association, at the scene level and aggregated to a summary for the entire asset.
- Scores are normalized to a scale of 0-100.
- General guidance is as follows, but custom rules can be applied as necessary:
- Scores of 30 or greater for IAB, places, activities, and objects show high correlation.
- For affinities, a score of 50 or greater relates to a high level of confidence.
- For GARM, it is recommended to use the risk rating (H/M/L) as the key indicator for triage.
Additional technical considerations and features:
- We use a RESTful API that conforms to the constraints of REST architectural style and allow for interaction with RESTful web services.
- The API utilizes a synchronous response structure. We confirm the receipt of a request with standard http error handling, and provide results back to a location of your choice tied to the scene.
- Custom taxonomies can all be supported.
- The delay to push JSON file into the system of record is 200ms.