Tech
OPPO drew attention with Artificial Intelligence Phones at CVPR 2024.
OPPO attracted attention with its artificial intelligence-supported phones at the 2024 CVPR conference. Standing out in the industry with its innovative technologies, OPPO continues to shape the future trends.
At the CVPR 2024 (Conference on Computer Vision and Pattern Recognition), a major international event, OPPO attracted attention by showcasing innovative artificial intelligence phone features. The company stated that it has been making intense efforts to develop AI-supported phones and reach a wider audience.
The flagship product showcased, OPPO Find X7 Ultra, stands out with its AI features. The device draws attention for having the computational photography system called OPPO HyperTone Image Engine. OPPO Find X7 Ultra claims to take high-quality photos with minimal computational area.
OPPO announced that with the establishment of the OPPO Artificial Intelligence Center at the beginning of 2024, it focuses its resources on enhancing AI capabilities. In June 2024, OPPO officially announced its strategy to bring AI phones to a broader audience, aiming to equip approximately 50 million OPPO smartphones with AI features by the end of the year.
Billy Zhang, OPPO’s President of Overseas Marketing, Sales, and After-Sales Services, emphasized that they participate in CVPR every year and particularly achieved significant milestones in 2024. Zhang stated that AI is a significant investment area for the company. He highlighted that AI phones make the user experience more innovative and CVPR is an important platform that underscores OPPO’s commitment to AI technology.
OPPO Research Institute and R&D teams selected 11 papers to present at CVPR. These papers cover research topics such as image restoration, digital human creation, video segmentation, dynamic motion capture, dynamic scene image synthesis, multi-view 3D object detection, and 3D creation.
Additionally, OPPO introduced UltrAvatar during CVPR, a method for creating drivable 3D virtual humans. UltrAvatar can generate drivable 3D virtual humans from a single image or text line using a disentangled color extraction model and an originality-guided texture mapping model.
Another significant contribution is UniVS, which presents the industry’s first universal large model and is used for video segmentation. This model allows segmentation of all objects in a specific category, meeting various real-world video segmentation needs.