Loading video...
Video Failed to Load
๐ข๐ข๐ขData release: high-res, multi-view, OLAT face recordings ๐ข๐ข๐ข We captured individuals in our custom light stage with 16 high-end, global shutter cameras (72 fps) and 40 LED modules, totaling 2.8M precisely calibrated frames. We us the data for BecomingLit (#NeurIPS2025): intrinsically decomposed Gaussian avatars, enabling photorealistic and real-time relighting... show more
19,536 views โข 10 months ago โขvia X (Twitter)
5 Comments

Aleksander Holynski9 months ago
why are these so creepy

halloichbins9 months ago
watching this on acid will probably fry your brain

Henry Pearce9 months ago
Great release, could you share information on the colour calibration process? Is the black level correct and consistent, as something doesnโt appear quite right

mrkelly9 months ago
The real cost isn't the cameras. It's the 18 months of calibration nobody sees.

Gaurav Sarkar :)9 months ago
Scary dataset
