High-Resolution Volumetric Reconstruction for Clothed Humans
DescriptionWe present a novel method for reconstructing clothed humans from a sparse set of, e.g., 1-6 RGB images. We revisit the volumetric approach and demonstrate that better performance can be achieved with proper system design. The volumetric representation offers significant advantages in leveraging 3D spatial context through 3D convolutions, and the notorious quantization error is largely negligible with a reasonably large yet affordable volume resolution, e.g., 512. Extensive experimental results show that our method significantly reduces the mean point-to-surface (P2S) precision of state-of-the-art methods by more than 50% to achieve approximately 2mm accuracy with a 512 volume resolution. Additionally, images rendered from our textured model achieve a higher peak signal-to-noise ratio (PSNR) compared to state-of-the-art methods.
Technical Papers Fast Forward Presenter
Event Type
Technical Papers
TimeTuesday, 12 December 20239:30am - 12:45pm
LocationDarling Harbour Theatre, Level 2 (Convention Centre)
