There’s gold in your video archive.
Online video viewing is exploding, creating extraordinary demand for archival video that can be repurposed to make new content. You can improve monetization of your video archive if you can quickly identify and serve up what people are looking for.
The capabilities of Artificial Intelligence to analyze video continues to grow with the development and enhancements of deep learning computer vision architectures like U-net, RetinaNet, SSD, YOLO, and Faster R-CNN. What's possible with AI Video Analysis? See below
An advertising agency needs aerial footage of the Rio de Janeiro coastline, shot specifically at sunset. Rather than shooting new footage, the agency can search and retrieve video from your archive that has been described and meta tagged by our AI engine, which can review your video archive and describe details like weather, landscape, objects, or shot selection, and historic landmarks.
People and Context
It’s crunch time and your news team needs video of a specific historical figure, speaking to a specific person at a specific event. Our AI Engine can recognize historical figures, identity whenever they appear in video, and describe their activities in each appearance so you can quickly search and find the videos you need.
Do you want to track how many times a specific player took a shot or dropped a pass? Use AI video analysis to track and record actions in live and taped sporting events, as granular as counting passes, shot attempts, and fouls. Data from past games can be tabulated and analyzed to enhance prediction models.
What makes viewers stay on a video longer? People can't explain what they really like, but AI can. With Computer Vision, we can capture visual elements nearly undetectable to the human eye, and then determine which elements are the ones that make people want to watch a video longer or click to the next video.
Ready to unlock the power of AI Video? Request a consultation.