VolHuMe dataset released with high-resolution 4D human scans

Researchers have introduced VolHuMe, a new dataset featuring high-resolution 4D human scans captured using 64 RGB and 32 depth cameras. This dataset includes detailed ground truth for 104 subjects, such as SMPL-X models, high-resolution meshes, and garment segmentation. VolHuMe's close-range, high-resolution capture method aims to preserve fine-grained body-part details, enhancing geometric fidelity and texture resolution compared to previous datasets. The researchers have also benchmarked current 3D and 4D human reconstruction methods using VolHuMe, highlighting the dataset's quality and the limitations of existing evaluation frameworks. AI

IMPACT Provides a new, high-fidelity dataset for advancing 4D human reconstruction and related AI research.

RANK_REASON The cluster describes a new dataset release for research purposes, detailed in an arXiv paper. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Hugging Face
• Niccolò Bisagno
• SMPL-X
• VolHuMe

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →