Most of the innovations when it comes to image recognition use AI technology and machine learning. Image recognition startups in Korea are looking for ways machine learning can take in data to not only make predictions but interprets the data, especially when it comes to texts, images, and videos. Image recognition is vital when it comes to self-driving cars which need to tell the difference between signs and people. This technology can also be applied for facial recognition as well. However, the biggest trend at the moment is how image recognition can be applied to augmented reality and the metaverse for retail operations. For example, there are many companies not just in Korea but all around the world that can use image recognition technology to see how often their products appear on social media platforms.
Image recognition powered by AI can be applied to many other industries. This revolutionary technology goes far beyond our imagination and there are some fantastic startups in Korea utilizing image recognition in innovative ways.
Top Image Recognition Startups in Korea
NOTA AI
Korean AI startup NOTA AI specializes in AI and machine learning to create a soft keyboard technology to reduce typos. In addition, they have developed a model compression technology that can be applied to mobile, IoT, drones, and robotics. The issue with AI is its high server cost and latency issues. Therefore, NOTA’s AI solution solves these issues with its core model compression technology. Therefore they work well for real-time object detection as well as detecting human emotions in real-time.
NOTA is supported by KAIST which is the top technical institute in Korea. Their focus for 2024 will be to research new solutions such as image segmentation and object recognition. They have raised a total of $23 million to date. Investors include Stonebridge Ventures, LB Investment, LG CNS, DS Asset, Company K Partners, Intervest, and Samsung SDS.
GDFLab
Korean AI startup GDFLab is an AI-based video quality enhancement company that increases the resolution and quality up to 4x the size and to 4K UHD. In addition, they use AI to develop IT solutions for various purposes such as image and video upscaling, recognition, and compression technologies. Their app, pikaVue is an AI-deep technology-based image and video Upscaler. It allows users to Upscale images up to 4x and restore images to high resolution. AI technology is used to eliminate noise and restore details that were not visible in the original image. Furthermore, users can convert their low-definition videos to FHD (1080P).
“Our AI-based real-time video quality enhancement quality would allow video streaming much smoother and faster. Our technology can save a lot network time and costs and can be used for surveillance cameras or video streaming services such as Netflix,” said the CEO of GDFLab, Kang Kyoung-ik.
GYnetworks (ReCon)
GYnetworks offers Video Analysis modules using Deep Learning for APIs, and SDKs in real-time. It can analyze all forms of video devices from CCTVs to personal video devices such as smartphones. They offer a video/image recognition solution called ReCon which can analyze the data continuously. This means it can offer 24/7 video surveillance analysis to offer users maximum security. It can be used for elevators, kid’s cafes, factories, labs, and even homes.
The advanced ReCon algorithm recognizes the movements that are important and ignores any activity that isn’t relevant to the scene. Currently, it is being used in hundreds of applications for detecting objects and authenticating users on still images and video streams in real time.
MARQVISION
Marqvision is an image recognition startup in Korea/US that aims to solve the counterfeit problem by using the latest AI technology. They were part of the Y Combinator accelerator program in 2021 and have focused mainly on IP protection using their technology. Moreover, they launched anti-piracy software that focuses on the gaming and entertainment industry.
Marqvision offers real-time counterfeit goods detection powered by AI. Brands from all around the world can automatically detect and remove infringing items from online stores/marketplaces such as Amazon. Now the process of detecting and reporting counterfeits can be fully automated. Their detection accuracy stands at over 95% and can also be applied to social media platforms such as Instagram.
