Twitter CEO Jack Dorsey claims the social network intends to use Magic Pony’s tech to enhance its strength in images and live video, opening up “a whole lot of exciting creative possibilities.”
The Magic Pony website now carries a Twitter logo and a statement announcing the acquisition. In terms of a description of the services it offers, it states that it has filed over 20 international patents over the past 18 months in the field of machine learning visual processing for web, desktop, and mobile.
Most importantly for Twitter, the startup’s tech boosts image and video quality by feeding media into a computer, converting it to a low-resolution, and then identifying the difference between the two sets of visuals. Additionally, Magic Pony has previously demonstrated how its system can improve pixels in live game streams in real-time (an area that we understand Twitter is interested in).
“Online video-streaming businesses rely heavily on video compression,” Magic Pony CEO Rob Bishop told MIT Technology Review in April. “Our first product demonstrates that image quality can be greatly enhanced using deep learning, and fast mobile GPUs now allow us to deploy it anywhere.”
Bishop added that his company’s tech could also improve the quality of images captured on smartphones with low-resolution cameras or in low-light. Consequently, it is easy to imagine Magic Pony assisting Twitter with improving the resolution of Pericope broadcasts, or general image and video uploads across its flagship platform.
Live-streaming has become a fierce battleground for Twitter, and Facebook. The latter’s Live feature is currently dominating headlines, for all the the right and wrong reasons. Perhaps Twitter believes it can gain the upper hand over its rival by boasting better image optimization tools and high-quality videos.
This latest purchase is Twitter’s third buyout in relation to machine learning startups, having previously acquired Madbits in July 2014, and Whetlab in June 2015. Magic Pony’s team will be integrated into Twitter Cortex; a team of machine learning experts dedicated to refining Twitter’s core product.
“Machine learning is increasingly … powering much of the work we’re doing to make it easier to create, share, and discover the very best content … every time you open Twitter,” Dorsey said.
