Defining True Video Comprehension (TVC). A ground-breaking, new concept within Computer Vision (CV) that applies to the field of content comprehension
How-to-Guide: 13 lines of code to integrate Netra and empower your business with Computer Vision
Transform your business with Computer Vision using Netra's APIs
Netra’s API can be implemented with only 13 lines of code. Because of its ease of implementation, there are several ways the output can be incorporated into strategies for leveraging visual content.
How our API works
Our API works under an asynchronous, request-response structure. For each video asset, we return a JSON file that clients can parse for targeting, analysis, or research purposes.
Three ways our partners are implementing our API in today’s media and marketing environment.
1. Publishers increase revenue by injecting brand safety and contextual classification in the bid stream via ad servers: here media owners initiate direct insertion orders based on Netra’s content scores:
- The advertiser establishes thresholds based on its objectives (e.g. relevant topic is “basketball” >=80 and brand safety >=95) and sets these requirements with the ad server.
- The publisher uses the Netra API to classify its content and relays this information to the ad server via ad tags or warehouse databases.
- When the publisher’s content achieves the advertiser’s matching thresholds, the direct deal ad placement opportunity is validated via the ad server. Ads are then served as part of the insertion order.
There are two key benefits when Netra’s scores are introduced into direct insertion order deals.
- First, for publishers, a standard taxonomy with validated scores is relayed to the ad servers’ deal requirements helping increase the value to the buyer of the inventory sold.
- Second, advertisers can set matching criteria given the standard taxonomy that enables easier, rules-based matching of creative to content to direct insertion orders. Given that both sides can ensure higher quality matches, the publisher benefits from premium pricing opportunities and the advertiser from higher quality targeting.
2. Integration with an SSP or DSP through PMP deals: similar to direct insertion orders, PMP deals are established based on mutually-agreed Netra scoring thresholds:
- PMP deals include an agreed buy-side and sell-side match on required Netra scores, e.g. both the content and the creative must have matching scores of >=80 for “basketball.”
- The publisher’s ad server confirms the match to ensure the PMP is executed under the required agreements.
The critical benefit resides in the assurance that the premium pricing under PMP deals is realized by ensuring a guaranteed, qualified match. Additionally, publishers and media owners are able to broaden their content offerings to new or increased volumes of PMP deals given that they are providing guaranteed, authenticated-quality content via Netra’s technology to advertisers.
3. Via video players: Netra’s technology can be implemented directly with video content management platforms.
- The video platform implements a call to Netra’s API and JSON file, and this information is returned to an internal database to characterize the video asset.
- After parsing the JSON file from the Netra API, the video platform relays the Netra scores to ad servers for premium pricing based on contextual, brand safety, and a host of other classifications
With Netra’s technology, video asset owners are able to achieve greater monetization. Brands also benefit from a greater depth of analysis and insight into video content performance.
A robust solution to the Post cookie world: PII is protected
Because content metadata is aggregated and scores are returned at a content level, PII and user-level data are protected, yet the quality and relevance of positioning content are preserved for brands. Both buy-side and sell-side partners are able to achieve appropriate monetization and audience reach with greater reliability based on complete content comprehension, including scene by scene deep comprehension of video content (versus relying on scraping potentially shoddy metadata descriptions of the video)
Netra’s technology includes a full comprehension of the text and imagery on the page and extends further through our Computer Vision (CV) technology into the full depth of context within video assets through frame-by-frame analysis on content detection, brand safety, emotion, and affinity.
Our vision is to empower the industry to unlock the untapped value within their content portfolios, provide relief and structured data classification, and empower media stakeholders to harness the massive amount of content created with each hour to turn it into an advantage…
If you are a media owner looking to maximize monetization in this period of transition to privacy-first activation, please reach out to us. We work with some of the biggest names in the industry and would enjoy speaking about how we can support your goals. If you are a brand looking to understand this new environment better, please get in touch. We can forward you to our publisher and vendor clients so you can test the benefits of buying with high fidelity classification with trusted partners.