In the evolutionary path that began with 2D photography and led to 3D modeling, reconstructing the light field alongside the matter field is the next great leap. Generalized Scene Reconstruction using Quidient’s 5D technology is the Final Frontier of Photography.
Generalized Scene Reconstruction (GSR) requires more than incremental changes to existing approaches. We reimagined a solution from top-to-bottom through the convergence of three key technologies.
Generalized Scene Reconstruction (GSR) requires more than incremental changes to existing approaches. We reimagined a solution from top-to-bottom through the convergence of three key technologies.
3 Key Technologies
Light Field Radiometry
Light transport theory describes how a given material emits, absorbs, reflects, scatters or transmits various frequencies of light. Quidient decouples the light field from the matter field by using transport theory to represent materials such as metal, wood, glass and even fog. This decoupling leads to significant advantages in machine learning / AI as well as scene compression and processing speeds.
Polarimetric Imaging
Traditional sensors capture the visible color and brightness of a light wave interacting with a material. Polarimetric cameras also capture the light wave’s non-visible “rotational orientation.” Quidient’s solution can interface with today’s newest polarimetric cameras. These cameras provide advantages in advanced GSR, especially when dealing with featureless surfaces and low lighting conditions.
The Database
Quidient’s Plenoptic (5D) Database technology represents a major shift in underlying architecture. It provides a novel means for separately encoding a matter field (as 3D Voxels, shown in turquoise) simultaneously with any light field (as 2D solid-angle elements, or “Saels”, shown in yellow). This spatially sorted, hierarchical approach leads to randomly accessible and searchable scenes with exceptionally fast subscene insertion and extraction. Its speed meets critical requirements for representing scenes with virtually unlimited levels of detail such as an interactive city map.
The Codec
Quidient’s Codec (5D) can encode and transmit some next portion of a Scene (i.e., “a subScene”) to be viewed on a remote device. Just as video codecs enable video streaming on common devices like smartphones, the 5D Scene Codec enables easy streaming of light field and matter field information.
Subscene Streaming
5D scenes of the future will be captured by potentially hundreds of cameras, each from a different perspective, even at different times under different lighting conditions. Whether real-estate for sale or a cityscape for virtual touring, these scenes of virtually unlimited levels of detail require a new “subscene” approach. Quidient’s Subscene Streaming technology is specifically designed for perspective agnostic, non-sequential subscene insertion and extraction, distributed over virtually any number of producing and/or consuming devices.
Brought to Common Devices Via Hardware Acceleration
The GPU was created to improve 2D image and video processing. Quidient envisions an entirely new approach for the low-level processing of spatial data. This new Spatial Processing Unit (SPU) offers several orders of magnitude improvement over the GPU. It will enable real-time encoding and decoding of large, shared scenes, allowing users to capture and manipulate scenes in the palms of their hands.
Physics-Based Machine Learning
Traditional “Black Box” AI/ML using conventional images requires massive training data sets and has not yielded adequate accuracy for the kinds of transformational applications that Quidient will enable. Quidient engines make extensive use of AI, including a novel ML approach called Physics-based Machine Learning.
Within the Engine:
Make Scene Reconstruction More Efficient
In Engine-Based Apps:
Create a powerfully accurate source of data for higher level AI tasks by feeding a 5D model into the ML for training (rather than a conventional image).