New AI method creates animatable 3D human avatars from single images

Researchers have developed DiGS-Avatar, a novel method for creating animatable 3D human models from a single image. This approach reformulates the task as a UV-latent completion problem, leveraging diffusion models for enhanced generative quality and a teacher-student framework to ensure 3D consistency. DiGS-Avatar reconstructs detailed 3D avatars with high fidelity and generalization capabilities in under a second. AI

IMPACT Enables faster and more accurate creation of 3D human avatars from single images, potentially impacting virtual reality and content creation.

RANK_REASON Publication of a research paper detailing a new AI method for 3D human reconstruction. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• 3D Gaussian primitives
• 3D human reconstruction
• arXiv
• Diffusion Models
• DiGS-Avatar
• teacher-student framework
• UV-latent completion
• UV-Space Diffusion

• paper
• product

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