In development
AI video outpainting
Extend video into the wider field of view that dome and panoramic screens require.
Explore productProducts & developmentAI imaging for
AI imaging for
immersive media.
Video outpainting, volumetric reconstruction and spatial playback software, supported by camera-system engineering and a full-scale display venue.
Research & development
Volumetric reconstruction
Reconstruct moving subjects from camera views, with depth and geometry that hold together through a performance.
Explore product
Licensed browser deployment · WebGPU & Vision Pro in development
Mirage
Spatial playback software, licensed for SuperMuseum: America. Now developing WebGPU and Vision Pro players.
Explore productSpatial intelligence researchImage geometry
Image geometry
and depth.
Our research connects real camera views, estimated depth and synthetic scene data. Volusynth explores multiview datasets with linked images, depth, camera calibration and pose information.
See the research programmeOne view. Linked spatial information.Data schematic
01RGB
Image & frame identity
02Depth
Distance at each pixel
03Camera
Calibration & pose
intrinsics
rotation
translation
1 imageCapture technologyControl the capture.
Control the capture.
Understand the geometry.
Portable rigs and studio arrays record synchronized views for stereoscopic, panoramic and volumetric processing. Optics, camera calibration and synchronization determine what the reconstruction software has to work with.
View the capture systemsGPU processingInference, reconstruction
Inference, reconstruction
and rendering.
Our development uses NVIDIA GPUs for image processing and spatial media. HP’s case study documents local AI inference, custom machine-learning workflows, background inpainting and high-resolution masking.
Current Mirage development extends GPU rendering into the browser through WebGPU, alongside native Vision Pro playback.
8 imagesSee what we’re building.
Ask about the demonstrated platform, current development or a technology partnership.