Korean startup GDFLab making it easier to upscale, compress & transmit high-quality videos using AI

GDFLab's AI technology helps in the image and video upscaling, recognition, and compression.

Transferring high-resolution videos or viewing it seamlessly without buffering can be a tiring process and costly many a time. Korean Artificial Intelligence technology company GDFlab has invented a solution so that everyone can enjoy high-quality videos without worrying about buffering or data cost.

Cost effective, seamless method for videos & image transfers

GDFLab or as the company terms it ‘Good Friend Artificial Intelligence Solution’ was founded in 2018. The startup uses AI technology to develop IT solutions for various purposes such as image and video upscaling, recognition, and compression technologies. Its product ‘GDF AI upscalar’ is an AI technology-based innovative video handler that dramatically increases video quality while reducing network and storage costs. The AI technology can save at least 80% of network traffic costs for delivering HD videos through the internet.

With the increase in internet traffic especially for video content, GDFLab’s upscalar ‘GDF Play’ will help clients who have to spend a lot of money for the network, storage, and server maintenance cost. GDF AI upscaler receives video contents with various format and protocol and generates better quality videos. For example, FHD input video can be upscaled up to 16K video while 360p input videos could deliver FHD video in the mobile app through embedded GDF AI upscaler.

GDFLab’s upscale technology can be used for compression & decompression for video on demand, which can be compressed up to 80% on the AI platform. It can be then upscaled into original resolution. For example, FHD 1080P video can be compressed into 270P video, and it can be decompressed into FHD 1080P again. This reduces the cost of transferring by 80%. This is particularly useful for mobile players where users can watch videos with no image quality deterioration

GDLab has also developed ‘PopPixel’, the AI platform that can upscale images up to 4X using a convolutional neural network and restores images to high-resolution quality using deep learning techniques.

License-based revenue model

The startup is having B2B traction performing Proof of Concept with COMMAX, a 50-year-old local company specializing in smart home and CCTV systems. GDFLab’s revenue model through a server license fee that is $ 20k per year. A user license fee of one dollar per use for over 100k subscribed users will be charged. The charge for maintenance services is $30k per customer per year for AI upscaling, SW upgrades, and bug fixes.

The company’s core team has over 10 years of experience with expertise in AI, Multimedia, and Security. The startup’s vision is to make video viewing better for everyone. GDFLab’s team will be attending the RISE Conference to be held from July 8 to 11 in Hongkong, that is supported by the Korea Institute of Startups & Entrepreneurship Development (KISED). The RISE Summit attracts many prominent leaders from global companies, major media outlets, and influential investors.

About KoreaTechDesk

KoreaTechDesk is a trustworthy source for news on the start-up industry in South Korea. It focuses on various tech start-ups, which are developing some unique and interesting products. press@koreatechdesk.com

View all posts by KoreaTechDesk →

Leave a Reply

Your email address will not be published. Required fields are marked *

20 + 3 =