Metaverse

Metaverse

The small vision metaverse technology includes virtual anchor, 3D facial reconstruction, dense facial landmark detection, human post estimation, sensitive motion capture and 3D scene perception. The technology has a huge application prospect and can be widely used in live broadcasting, banking, government affairs, gaming, entertainment and other scenes. It will create a rich-function digital avatar product that can deeply interact with real or virtual scenarios and deepen metaverse applications.

Metaverse

  • Virtual Anchors

    Virtual Anchors

    Using 3D motion capture technology, multiple 3D joint points are captured in real time and their joint poses are computed. The captured pose data is used to control the interaction of the avatars in metaverse space.

  • Deep 3D Reconstruction

    Deep 3D Reconstruction

    With the advent of technologies such as digital twins, the development of deep 3D reconstruction technology is used for 3D modelling of objects, buildings and large-scale scenes, which is further used for rendering 3D visualization of digital cities, 3D display of factories, smart factories and so on.

  • 3D Facial Reconstruction

    3D Facial Reconstruction

    Use 2D images to finely reconstruct 3D images with over 50,000 facial key points, using 3D models and adding interactive elements in the graphics engine.

  • Dense Facial Landmarks

    Dense Facial Landmarks

    Facial landmark detection is the process of locating the position of key areas of a particular person's face, including eyebrows, eyes, nose, mouth, facial contours, etc., and is an important and fundamental part of the face recognition task.

  • Human Pose Estimation

    Human Pose Estimation

    It accurately extracts the joint points of the human skeleton and constructs the human skeleton model, providing the basis for further applications such as human pose recognition and real-time interactive gaming.

  • Sensitive Motion Capture

    Sensitive Motion Capture

    Human skeleton, hand and 3D facial key points are accurately extracted to drive the 3D model to perform the appropriate movements in real time.

  • 3D Scene Perception

    3D Scene Perception

    AI models that perceive 3D space by estimating the depth of the scene with a low-cost regular camera implement high-quality photo and video enhancement, including enriching content details, adding special effects and enabling AR capabilities.