Xvisio Technology (Shanghai Xvisio Sensing Technology Co., Ltd.), a Silicon Valley-based startup founded by overseas returnees, has for many years focused on core technologies and products in the field of spatial perception and interaction. From core technology research and development to ODM services and product platform development, Xvisio Technology has accumulated over 20 core IPs, achieving independent control and comprehensive integration across algorithms, sensors, hardware platforms, application SDKs, and system optimization. By collaborating with industry partners, Xvisio Technology has enabled numerous industry customers to build XR systems and ecosystems, playing a key role in the development of the Metaverse ecosystem. At the World Internet Conference in September of this year, Xvisio Technology's SlimEdge E34RB, an enterprise-grade AR glasses, garnered significant industry attention.

At the World ARVR Industry Conference in October, Xvisio Technology launched a number of new AR HMD products, and teamed up with Epson at the 4th China International Import Expo to launch AR guide solutions for the cultural and tourism industry, completing a successful practice of digital cultural and tourism and actively exploring the implementation of metaverse technology.

So, what enables Xvisio Technology to continue to make progress on the journey to the metaverse?

1. High-performance VSLAM technology is the basis of spatial perception interaction

VSLAM (Simultaneous Localization and Mapping) is the core of achieving a 6DOF immersive spatial perception and interactive experience. Along with near-eye display, VSLAM is considered one of the two pillars of XR. While VSLAM technology has a long history in academia, it has only been implemented in production in the past two years due to limitations in computing power within product platforms. Xvisio Technology excels in the industry for its high-performance VSLAM technology and solution deployment. Its unique approach utilizes independent VPU/DSP computing units to handle computing and sensor fusion tasks, improving system performance and design flexibility. Microsoft Hololens, Magicleap, and the rumored upcoming Apple MR headset all utilize a similar architecture to ensure performance. Xvisio Technology achieved device-level deployment of this architecture as early as 2018, earning recognition as a "Key Technology Breakthrough" enterprise by the Ministry of Industry and Information Technology that same year. Continuous iterations in recent years have resulted in its deployment on multiple chip platforms. At this year's World ARVR Industry Conference, Xvisio Technology unveiled a product solution supporting binocular and quad-camera VSLAM, boasting accuracy comparable to HoloLens. To date, Xvisio Technology has empowered numerous industry clients, including several industry leaders, with its ODM services and corresponding platform products, based on its core perception and interaction technologies.

SlimEdge B50R Pro AR glasses based on four-eye VSLAM

2. Full-chain underlying technology supports a rich product platform

To achieve an immersive interactive experience, the accuracy, speed, and dimensionality of spatial perception interaction (gestures, controllers) are crucial foundations. Currently, most product developers lack such comprehensive system integration, comprehensive R&D, and product deployment capabilities, which limits the widespread adoption of AR/MR products. Over the past three years, Xvisio Technology has continuously iterated its native VSLAM algorithm, enabling deployment on mainstream hardware platforms such as Intel, Qualcomm, and MTK. It has also developed full-link low-latency multi-rendering optimization technology and integrated it with a full range of optical near-eye display solutions, forming an XR system architecture based on a spatial intelligent perception layer and an interaction layer. Xvisio Technology's XR perception layer, built on a standalone chip, enables on-the-edge processing and fusion of multiple sensors, including binoculars, quad-camera fisheye sensors, Time of Flight (TOF), RGB, infrared cameras, low-light cameras, and IMUs. It provides localized processing capabilities for high-speed VSLAM, high-speed depth detection, object pose detection, object recognition, and audio and video interaction. This greatly facilitates the modular design and deployment of the system, making it suitable for multi-modal spatial perception interaction scenarios, including XR terminals, wearable devices, and robots. The interactive capabilities for application scenarios are realized by the Xvisio AR foundation SDK combined with the perception layer capabilities, providing the main functions for interacting with XR, including spatial anchoring, plane detection, map creation, virtual and real integration, 3D reconstruction, gesture interaction, multi-person collaboration and end-cloud integration. It supports mainstream operating system platforms such as Android, Windows, and Linux, enriching the entrance to the metaverse.

3. Precise spatial anchoring and integration of virtual and real

The integration of virtual and real is the core value of AR/MR devices in the Metaverse. It is the gateway to the Metaverse, yet also the most challenging aspect. Achieving precise virtual-real integration and spatial anchoring requires not only precise VSLAM technology but also high-precision optical perspective system calibration and low-latency system optimization. Xvisio Technology has developed a complete technical system and workflow for high-precision positioning and virtual-real overlay in this field, supporting deployment on all-optical hardware platforms (PC, Android, and Linux). This accumulated expertise provides the industry with diverse technical options and broadens the entry point into the Metaverse.

4. Open up your phone's "Ren and Du meridians" to enable a smooth 6DOF experience

Mobile phones, as the gateway to the mobile internet, have a large existing market. AR glasses, as mobile phone accessories, have long been considered the best application scenario for XR. However, many AR glasses on the market are still mainly used for screen projection and viewing. The reason is that to obtain a smooth and immersive 6DOF experience, the current mainstream solutions require modifying the underlying code of the mobile phone to achieve low-latency processing, and require mobile phone manufacturers to open root permissions. This is undoubtedly a huge challenge for general-purpose AR glasses. Xvisio Technology, based on its unique end-processing split architecture, achieves low-latency processing on non-rooted mobile phones, achieving the same experience as customized devices. This technology breaks the deadlock in the implementation of 6DOF AR glasses in the mobile phone market, can quickly drive the content ecosystem based on mobile platforms, and will greatly expand the application scenarios of the metaverse.

5. Head-hand collaboration with 6DoF interaction enhances the immersive experience

In terms of immersive interaction, AR and VR both require 6DOF head tracking. The difference is that AR focuses on understanding the environment, while VR emphasizes acquiring spatial information. Regarding hand interaction, AR favors natural gestures, while VR favors controllers. 6DOF controllers have become standard for VR devices. Facebook (Meta) recently released a controller with built-in camera active tracking, based on an independent VSLAM computing unit, representing the industry's technological direction. Xvisio Technology launched a similar product design as early as 2018 and has been granted a US patent. Xvisio Technology also supports CV-based 6DOF controller solutions, and a quad-camera-based 6DOF head-hand solution is already in the process of being commercialized.

In the field of natural gesture interaction, Xvisio Technology has developed high-speed gesture recognition technology based on AI algorithms, and its product platform is also compatible with other gesture solutions on the market.

6. Multi-person collaboration, end-to-end cloud integration, and cross-temporal interaction

The metaverse will transform our work and living spaces. Cross-temporal collaboration will be a new interactive capability brought to us by XR devices. Rapid environmental understanding and 3D spatial reconstruction are crucial foundations for cross-temporal collaboration. Traditional 3D reconstruction requires two stages: scanning and post-processing. Xvisio Technology, leveraging its powerful edge-processing sensor fusion and VSLAM technology, has achieved real-time 3D reconstruction, map creation and sharing, scene recognition, and end-cloud integration, laying the foundation for multi-person remote collaboration. It also provides a rapid deployment platform for virtual simulation and digital twins. At the 2021 China International Import Expo (CIIE), Xvisio Technology demonstrated cloud-rendered 3D virtual sandbox technology and end-cloud-based immersive 3D AR navigation and scene recognition technology, completing the implementation of end-cloud-integrated large-scale scene navigation using lightweight AR devices. 5G-based high-bandwidth, low-latency cloud rendering technology enables re-rendering of complex multi-scene models in the cloud, unaffected by system computing power. This allows multiple people to share and interact with cloud-rendered content, enhancing the interactive experience across time and space.

7. Immersive Volumetric Video Content Generation

Leveraging powerful on-device processing and sensor fusion capabilities, Xvisio Technology's SlimEdge XR platform provides RGB video streams with high-precision synchronized depth point clouds and pose information. This platform can be used for volumetric video capture, rapid XR content creation, and 3D object reconstruction, facilitating the creation of immersive user-generated content (UGC) in the metaverse era. Creators can use head-mounted XR devices for first-person perspective capture or Xvisio Technology's standalone camera for third-person perspective capture.

8. Core technology platformization facilitates the implementation of Metaverse applications

Since its inception, Xvisio Technology has adhered to a development path that integrates algorithms and hardware, from underlying sensors to product platforms. While serving leading industry clients, the company has built comprehensive core technology and product deployment capabilities, launching a diverse portfolio of AR/VR solutions and machine vision modules. Currently, Xvisio Technology's core business models are customized ODM services based on core perception and interaction technologies for the XR industry, as well as standardized platform product delivery, to support the implementation of Metaverse applications.

"Innovating machine vision perception capabilities and surpassing the limits of human vision" is the shared vision of the Xvisio Technology team. As the concept of the metaverse continues to gain momentum, Xvisio Technology will work with all ecosystem partners to build a path toward the metaverse.

Source: VR Gyro Original Report Link: Building the Road to the Metaverse

Founded in Silicon Valley in 2016 and established in Shanghai after receiving angel investment in 2017, Xvisio Technology is a solutions provider specializing in spatial perception and interaction, providing underlying technologies and system integration for AR/VR headsets. Its core product, SlimEdge XR, is an integrated edge processing platform that integrates a VSLAM engine, a depth engine, an AI engine, an audio and video engine, and gesture and eye control, suitable for both standalone and standalone XR devices. The founding team, all from Fortune 500 companies and a series of entrepreneurial ventures, has accumulated years of industry-leading research results and currently has subsidiaries in the United States and Europe.