LOS ANGELES--(BUSINESS WIRE)--Video Gorillas, a developer of state-of-the-art media technology incorporating machine learning, neural networks, visual analysis, object recognition, and live streaming, announced today the commercially available version of its Bigfoot Frame Compare product, which is set to redefine the way film, television, and post-production companies manage assets, finishing, and remastering and preservation projects.
Bigfoot is a scalable, lightweight proprietary technology that leverages Video Gorillas’ patented computer vision/visual analysis, Frequency Domain Descriptor (FDD), and machine learning technology. It is designed to automate the manual-labor intensive conform process (which matches an original frame of film to the final edited work) and the compare process (which compares unique or common frames between different film cuts) by finding like “interest points” common across a series of images or frames of film.
Existing interest point matching methods require enormous computer processing speeds and large index sizes, making them difficult to scale. Similarly, pixel-matching solutions have limitations that make them impractical for all but very specific, uncommon film and television projects. Thus conform is traditionally a bottleneck in post-production because “eyematching” has been the only consistently reliable solution – a human being in an edit suite matching frames by eye. This effort requires highly skilled labor, expensive facilities and data storage, and the work is time-consuming, inefficient, and not creative. For example, a one hour, VFX-heavy television episode with 1.3 million frames (approximately 80,000 feet of film) and 89 scans with a duration of 14:56:20 (around 16TB) would ordinarily take 8-10 days to conform. By contrast, Bigfoot is a scalable, automated, lightweight solution capable of completing the same task in 10-12 hours.
“We worked with a select group of film and television customers on hundreds of projects for several months to prove the Bigfoot concept and achieved exceptional results, and we’re thrilled to make Bigfoot widely available to the Media & Entertainment industry,” said Video Gorillas CEO Jason Brahms. “Bigfoot enables a previously unimaginable level of speed, efficiency, and transparency in remastering, localization and restoration workflows, reducing the time to market for content owners, unlocking additional value in film and television libraries, and supporting ongoing preservation efforts.”
Bigfoot Frame Compare is capable of performing complex tasks at high speeds, such as differential analysis of frames that are unique, common, or have shifted or moved between two versions or cuts of a film or TV show, as well as determining whether the common frames are identical. Bigfoot can auto-conform restoration or remastering projects by comparing the frames from a reference picture to those from film scans, and reconstruct the timeline using the sequences of frames that have the most points in common. Additional conform projects Bigfoot supports include:
- A/B reels to reference picture conform
- Matching VFX plates or green screen to reference picture
- Trailer reconstruction from scans
- News reel reconstruction
- Matching stock footage
Bigfoot, which as of today is commercially available for companies across the Media & Entertainment industry, combines state-of-the-art machine learning technology and delivers it all inside Docker containers, making it the easiest way for developers to build and deploy artificial intelligence into their applications both in the cloud and on premises. It also enables front-end validation UIs that work in a web browser.
About Video Gorillas
Video Gorillas is a media-focused product and services company that develops state-of-the-art video technology incorporating machine learning, neural networks, visual analysis, object recognition, and live streaming. The company is headquartered in Los Angeles with engineering based in Kiev. For more information visit: http://www.videogorillas.com.
Copyright © 2018 Video Gorillas LLC. All rights reserved. Video Gorillas LLC, its subsidiaries, and their respective taglines are either trademarks or registered trademarks of Video Gorillas LLC or its subsidiaries in the United States and/or other countries. All other trademarks used are owned by their respective owners.