See how it works

reuters.jpg, 2 kB

Viewdle Advantage

“Recognize” the Difference

The Problem: Finding the exact video you want to watch, at the precise point you want to see it, without sorting through irrelevant clips or cuing footage

For example, when you search for a “Paris Hilton” video clip using existing search methodologies, hundreds of video clips will appear, but the majority of the results may not include an actual appearance of Paris Hilton on screen, may not be contextually relevant, and may not reliably showcase the point in the clip where she appeared. Instead, you may get clips where an on-screen entertainment news anchor is talking about Paris Hilton, where a YouTube user has added Paris Hilton tags to draw more attention to his video, or you may even get clips of Paris, France.

The Solution: Viewdle

Viewdle is a video indexing and reference system that fuses pioneering, facial-recognition visual analysis technology along with other leading search techniques to automatically look “inside the video” and identify true on-screen appearances with unrivaled accuracy and relevance. Viewdle’s frame-by-frame analysis provides search results based upon the actual video assets, not just the associated metadata. Viewdle solves the inefficiencies inherent in using text-based metadata and other existing tools for indexing, searching and retrieving moving media.

Using the example above, Viewdle users are not only guaranteed videos that actually show Paris Hilton on screen, they can also search for a specific appearance and jump to an exact section within the video (e.g. Paris Hilton leaving jail and getting into her limousine). This is the Viewdle difference.

Technology Highlights