This Demo is not designed for Mobile Devices.

Please view on high resolution monitor on Desktop.

If you see this in a desktop browser, expand the width of your browser. Chrome is Prefered.

If you are on a Govt. or Corporate network this site and the content may not display or playback correctly.  Reload the page if you have any issues.

please let page load fully – movie payload size 60-200mb 
use slider to resync videos

COMMON PHOTOGRAMMETRY

C/AR (Component Albedo & Roughness)

Both captured handheld (not turntable)

  • Ambient lighting, partial sun, 1pm
  • 38MP, 21mm Zeiss prime, handheld, 125th shutter, f/16, 800 ISO
  • No cleanup of model, original 243M tris decimated to 100K
  • Diffuse & normal map, twenty nine 8K UDIMs
  • Rendered unlit in UE4 at 4K
  • Polarized lighting captures empirical albedo & roughness values
  • 38 MP sensor, 21mm prime lens, handheld, 125th shutter, f 8.0 and 11.0, 1600 and 3200 ISO
  • No cleanup of model, original 797 million tris decimated to 1 million
  • Diffuse, roughness, and normal maps, fourteen 8K UDIMs, empirical-based isolated specular drives normal and roughness
  • Rendered lit (animated lighting) in UE4 at 4K

Common photogrammetry of aged asphalt returns high resolution, but lighting conditions are suboptimal. Valued data hides in the shadows (note the shift mid-capture). Even if you manage to hit the sweet spot, diffuse/not too diffuse, baked lighting means you’re either stuck with it or will pay dearly in labor changing it for lesser results.

If the ground truth in the common photogrammetry reflects the “true” color, grey asphalt, then why in this example does AC/R appear so warm? Because true for 1pm doesn’t stay true — sunset. Empirical component albedo frees AC/R to reveal a deeper truth, relighting being one way to reflect the very nature of change. (The truth goes yet deeper, see albedo color under Technical Study.)

Even common photogrammetry is gaining ground as a tool used for anomaly detection in infrastructure. Turbo-charged with AI much can be learned about deterioration by analyzing materials photographed in ambient light. The slider makes obvious how much information is missed with common photogrammetry.

The value of roughness information, in this case the glint from felspar in asphalt, may shine through with the right lighting conditions. Relightable AC/R makes evident the richness of roughness. What’s easily hidden in plain sight is more effectively revealed when switching off base color, then normal intensity, and lastly with animated lighting. Consider the value to anomaly detection of pointing an AI at AC/R training sets.

Common photogrammetry typically captures scenes with ambient lighting as we have done in this comparison. The result is that the lighting, including shadows and specular reflections, is “baked” or trapped in the photography.

AC/R records separately how light interacts with materials, so that lighting can be tuned, even animated, in any way imaginable after capture.