Skip to main content

🏢 Electronics and Telecommunications Research Institute

KOALA: Empirical Lessons Toward Memory-Efficient and Fast Diffusion Models for Text-to-Image Synthesis
·5238 words·25 mins· loading · loading
Computer Vision Image Generation 🏢 Electronics and Telecommunications Research Institute
KOALA: New efficient text-to-image diffusion models achieving 4x speed and 69% size reduction of SDXL, generating 1024px images on consumer GPUs with 8GB VRAM.
ContactField: Implicit Field Representation for Multi-Person Interaction Geometry
·3542 words·17 mins· loading · loading
AI Generated Computer Vision 3D Vision 🏢 Electronics and Telecommunications Research Institute
Novel implicit field representation accurately reconstructs multi-person interaction geometry in 3D, simultaneously capturing occupancy, instance IDs, and contact fields, surpassing existing methods.