17984521. REPRESENTING 3D SHAPES WITH PROBABILISTIC DIRECTED DISTANCE FIELDS simplified abstract (SAMSUNG ELECTRONICS CO., LTD.)
REPRESENTING 3D SHAPES WITH PROBABILISTIC DIRECTED DISTANCE FIELDS
Organization Name
Inventor(s)
Tristan Ty Aumentado-armstrong of Toronto (CA)
Stavros Tsogkas of Toronto (CA)
Sven Josef Dickinson of Toronto (CA)
REPRESENTING 3D SHAPES WITH PROBABILISTIC DIRECTED DISTANCE FIELDS - A simplified explanation of the abstract
This abstract first appeared for US patent application 17984521 titled 'REPRESENTING 3D SHAPES WITH PROBABILISTIC DIRECTED DISTANCE FIELDS
Simplified Explanation
The present patent application describes a method for representing shapes using probabilistic directed distance fields. This method involves obtaining a camera representation and a latent shape vector representation of a scene. The camera representation provides information about the position and direction of the view of the scene.
- The method calculates a visibility score and depth for each ray of a plurality of rays emanating from different positions and directions determined by the camera representation.
- Renders of geometric information of the scene are generated using the visibility score and depth of the rays.
Potential Applications
This technology has potential applications in various fields, including:
- Computer graphics: The method can be used to generate realistic renders of scenes, allowing for more accurate and detailed visualizations.
- Virtual reality: The technology can enhance the immersive experience in virtual reality environments by providing more realistic and visually appealing scenes.
- Gaming: The method can be utilized to create lifelike environments and characters in video games, improving the overall gaming experience.
Problems Solved
The technology addresses several problems in representing shapes and generating renders:
- Accuracy: The probabilistic directed distance fields provide a more accurate representation of shapes, resulting in more realistic renders.
- Efficiency: The method efficiently calculates visibility scores and depths for a large number of rays, allowing for faster rendering of scenes.
- Flexibility: The latent shape vector representation allows for easy manipulation and modification of shapes, providing flexibility in creating and editing scenes.
Benefits
The use of probabilistic directed distance fields for shape representation and rendering offers several benefits:
- Realism: The technology enables the creation of highly realistic renders, enhancing the visual quality of scenes.
- Speed: The efficient calculation of visibility scores and depths allows for faster rendering, reducing the time required to generate renders.
- Flexibility: The latent shape vector representation provides flexibility in shaping and modifying scenes, allowing for easy customization and experimentation.
Original Abstract Submitted
The present disclosure provides methods, apparatuses, and computer-readable mediums for representing shapes with probabilistic directed distance fields. In some embodiments, a method includes obtaining a camera representation and a latent shape vector representation of a scene. The camera representation indicates position information and direction information of a view of the scene. The method further includes calculating, based on the latent shape vector representation of the scene, a visibility score and a depth for each ray of a plurality of rays emanating from a corresponding plurality of positions and directions. The plurality of positions and directions are determined from the camera representation of the scene. The method further includes generating renders of geometric information of the scene using the visibility score and the depth of the plurality of rays.
