V-Nova Adds Machine Learning, New U.S. Customer

V-Nova told us at NAB 2018 that it now has multiple customers delivering live 1080p content at less than 1 Mbps. 720p content can be delivered at 450Kbps and 360p content at 120Kbps. The company also has its first U.S.-based customer: uCast. They are an OTT provider of content. The low bit rates, encoding efficiency and scalability of the Perseus Plus solution were key factors in the win, V-Nova told us. “They were able to spin up a channel in a day,” we were told. The platform powers services such as Qello Concerts, Stadium, and Acorn TV.

V-Nova also announced that the Perseus Pro codec has been upgraded to include machine learning capabilities. The Perseus Pro codec powers the P.Link Dual UHD/8HD video encoder and decoder, designed for contribution signal distribution.

Guido Meardi, CEO and co-founder at V-Nova, said:

“P.Link and the Perseus Pro codec that powers it had already revolutionized the contribution space by enabling studio-quality, intra-only HD and UHD at vastly lower bitrates than existing solutions. With our machine-learning advances, we’re now able to deliver visually lossless UHD at just 140Mbps, which is comparable with what many operators are currently using to deliver HD.”

Before the upgrade, Meardi told us the UHD signal required 200-250 Mbps for the same quality level.

By incorporating Machine Learning and more specifically, Convolutional Neural Networks into its multiscale approach, Perseus Pro has been trained to identify common video features such as edges and curves. By identifying those features across the various layers, the amount of residual data required for reconstruction is reduced and allows Perseus Pro to deliver an accurate picture, even at lower bitrates. This means huge efficiency gains over J2K and MPEG4 solutions, the company claims.

By 2020, it is predicted that video will account for roughly 80% of all Internet traffic, 20% of which will be UltraHD. This demand for higher definition video, delivered over existing network infrastructure, is placing more strain on operators and their network providers, who desperately need efficient codecs to avoid costly upgrades to their connections. – CC

Step 1 of 4