With the development of information technology, the application of holographic, AR, and VR in games, communication, education, digital twin, etc., is gradually increasing, and the reliability requirements for graphics rendering are getting higher. Rendering processing directly determines VR and AR image quality. Compared with traditional image rendering, VR and AR have higher rendering burdens and lower latency requirements in intense interactive application scenes (such as games). In response to the needs of WiMi’s cloud development business division, the R&D team has developed a highly reliable Graphics Rendering Engine Optimization System based on cloud computing. WiMi Hologram Cloud, Inc. (NASDAQ:WIMI) combined cloud rendering, AI, and foveated rendering based on cloud computing to optimize the rendering engine in several aspects, such as rendering resources, API level, memory and operation, CPU’s arithmetic power, the liberation of GPU pressure at CPU, etc. The system utilizes AI’s deep learning capability to achieve image noise reduction, anti-aliasing and rendering load reduction, etc.
WiMi’s Cloud Computing-based, Highly Reliable Graphics Rendering Engine Optimization System allows different content applications to be more easily adapted to differentiated terminals and solves complex user experience problems.
- Cloud rendering focuses on collaborative rendering at the edge of the cloud network, importing the rendering capabilities required by applications into the cloud, which can effectively reduce terminal configuration costs and enable users to experience better rendering effects.
- Eye-tracking technology can ensure that the reduction of rendering arithmetic and the number of display pixels does not reduce the user experience resolution. The foveated rendering technology is selected as the industry standard, and the foveated rendering and foveated optics based on eye-tracking are the current research hot spotsof the point technology architecture.
- Artificial intelligence is a booster of rendering quality and efficiency. Artificial intelligence can accelerate high-performance noise reduction processing and reduce high-fidelity image rendering time. Therefore, artificial intelligence will be a significant research direction for cloud dyeing configuration in different applications and network environments.
Foveated rendering is based on eye-tracking, which only renders the central area where the user looks with high precision and renders other areas within the field of view with low resolution, which can both reduce GPU shading load and ensure a high frame rate, thus better allocating the limited computing resources for a better user experience. Eye-tracking interaction has the advantages of low interaction cost and a short interaction path. With eye-tracking interaction, head movement only controls the field of view, and eye movement locks the target, making the user experience more relaxed and natural even in a more comfortable environment. User behavior analysis uses the information collected by eye-tracking manufacturers for developers, and they can then optimize and customize products, game production, etc.
With the development of cloud rendering, artificial intelligence, and foveated rendering, the balance between rendering quality optimization and efficiency will be further enhanced by researchers. WiMi has done a lot of research in this field, and the cloud-based graphics rendering engine optimization system will also provide users with better holographic rendering optimization cloud services.