Berlin-based Segmentive secures €2 million to launch new live video editing tools for mobile

Deep-tech computer vision startup, Segmentive, today announced closing its €2 million seed round in a bid to deliver whole-body video segmentation, in real-time and on mobile.

Founded in 2019, Segmentive allows users to intelligently remove themselves from live backgrounds to be transported into a scene of their choice, enabling a broad range of AR/VR use cases. Segmentive’s cutting-edge approach combines deep learning techniques with the industry’s most advanced online learning neural network algorithms to replicate and scale human understanding of scene identification. The startup is actually a spin-off from parent company, Cinector, with years of special effect experience in movies. 

To date, removing the background from a video of a moving person on a mobile device in real-time has been impossible. The technology is designed to be used by social media platforms looking for unique features to transport users to different worlds by changing their backgrounds. Segmentive has applications with mobile phone manufacturers and tech giants looking to push the boundaries of computer vision to create new commercial applications.

Lutz Roellig, Segmentive CEO commented: “[This] funding will accelerate the delivery of our multi-layer segmentation technology mobile prototype and allow us to make key hires to support execution. VR and AR use cases have been held back by an inability to display crisp edges on mobile devices as humans move around in virtual environments –  Segmentive’s technology breaks through this barrier to deliver fine details in real-time video on modern mobile devices.”

Segmentive unlocks new computer vision capabilities, never seen before by pre-classifying each frame into background and foreground using a pre-trained, convolutional neural network. Using online learned classifiers, Segmentive can achieve super-sharp edges in real-time dynamic videos.

The €2 million seed round was led by Leipzig based Technologiegruenderfonds Sachsen (TGFS). TGFS is an early-stage VC fund targeting high tech startups with strong growth potential and has invested in over 120 companies in its 20-year history.

TGFS’s Investment Director, Friedemann Stier commented: “Segmentive has proven the technology works and TGFS is pleased to be able to fund the growing team of computer vision experts to deliver the first working prototype. Leadership, combined with the teams’ skill and ingenuity makes Segmentive stand out as a winning combination.”


6 Startup Lessons From Editing Videos For A Shanghai Incubator

Deglorifying entrepreneurship, the burn out, team work, letting go, everything you wish you’d knew before starting as an entrepreneur.

Entrepreneur's Handbook – Medium

Vochi, the ‘computer vision’-based video editing and effects app, raises $1.5M seed

Vochi, a startup operating out of Belarus that’s created a “computer vision”-based video editing and effects app for mobile phones, has raised $ 1.5 million in seed funding.

Leading the round is Ukraine-based Genesis Investments (backer of BetterMe and Jiji). It follows pre-seed funding in April 2019 from Bulba Ventures, where Vochi founder and CEO Ilya Lesun previously worked as a product analyst.

Recognising that video content is booming, especially on short video platforms like TikTok, Lesun sort to build a mobile video app that would help creators stand out, while making creative video editing simple.

With the help of a computer vision-based video segmentation algorithm, the app lets you apply effects on single objects in your videos, opening up a lot of opportunities to make something unique. You can mix different styles and play with different scenarios, all in real-time, which helps visualize how the finished video will look, before downloading a high quality copy to begin sharing.

“As mobile content production and consumption increase, there’s also an increasing demand for content creation tools,” Lesun tells TechCrunch. “This is where we can bring value to content makers — be it professionals and amateurs — by delivering a video editing platform with as many tools and effects as possible. We see it as a content editing studio in a user’s pocket”.

To that end, its the use of computer vision and its ability to apply effects, stickers, and filters to any objects in a video, that Lesun says differentiates the Vochi app from competitors. “The app uses object segmentation algorithms that allow users to edit single objects in the video,” the founder explains. “[This lets users] to apply high-quality effects on single objects in 1080p videos in real-time”.

Typical users are cited as a “content creators”: people that share a lot of content on social media and “don’t like to keep their videos stored in the phone gallery”.

Adds Vochi: “It can be an influencer or blogger, who really wants the content to stand out and be really engaging for the audience. It can also be a creative person, who likes to create fun videos and share those with friends. And, of course, it’s basically every smartphone user”.

Startups – TechCrunch