Ido.Sobol
@IdoSobol
Excited about image-conditioned diffusion models like Zero123 but struggling with reliable outputs? Come visit poster #2404 at #NeurIPS2024 to learn how to elevate ‘Zero to Hero’!🦸♀️🦸♂️ I’ll be there to chat research and answer your questions! ⏳ Happening in two hours
We are presenting "Zero-to-Hero" today at @NeurIPSConf, join us to hear about improving diffusion models, by filtering the attention maps, without further training! 📅Today, 11 a.m. 📍East Exhibit Hall A-C #2404 zero2hero-nvs.github.io #NeurIPS2024 #NeurIPS
Your Diffusion Model May Know More than it Shows 🐢🦝 Excited to share "Zero-to-Hero" 🦸 to appear at #NeurIPS2024! We propose a test-time filtering method for attention maps that significantly enhances image-conditioned diffusion models. Kudos to @IdoSobol @Chenfeng_X 🧵 1/n
Symmetries are everywhere — from butterfly’s wings to Greek temples. But detecting them in noisy data? That’s a challenge. 🦋🏛 Our #SIGGRAPHAsia2024 paper, Robust Symmetry Detection via Riemannian Langevin Dynamics, tackles this: symmetry-langevin.github.io 🧵(1/n)
Your Diffusion Model May Know More than it Shows 🐢🦝 Excited to share "Zero-to-Hero" 🦸 to appear at #NeurIPS2024! We propose a test-time filtering method for attention maps that significantly enhances image-conditioned diffusion models. Kudos to @IdoSobol @Chenfeng_X 🧵 1/n