Tootfinder

Opt-in global Mastodon full text search. Join the index!

@arXiv_csCV_bot@mastoxiv.page
2025-09-26 10:26:01

Does FLUX Already Know How to Perform Physically Plausible Image Composition?
Shilin Lu, Zhuming Lian, Zihan Zhou, Shaocong Zhang, Chen Zhao, Adams Wai-Kin Kong
arxiv.org/abs/2509.21278

Does FLUX Already Know How to Perform Physically Plausible Image Composition?
Image composition aims to seamlessly insert a user-specified object into a new scene, but existing models struggle with complex lighting (e.g., accurate shadows, water reflections) and diverse, high-resolution inputs. Modern text-to-image diffusion models (e.g., SD3.5, FLUX) already encode essential physical and resolution priors, yet lack a framework to unleash them without resorting to latent inversion, which often locks object poses into contextually inappropriate orientations, or brittle at…

@midtsveen@social.linux.pizza
2025-09-28 21:04:00
Content warning: ⚠️ Autism, Special Interest, Zoë Straub ⚠️

I can spend days, even weeks here, because when my brain locks onto Zoë Straub in special interest mode, it doesn’t hold back, it goes all the way in. Since May 14, 2016, I’ve been drawn to Zoë Straub, that exact day when Loin d’ici grabbed me, the ESC version, and it’s never just been about the song, it’s not a simple idea floating around in my head, it feels like something carved deep inside, like a part of me that’s permanent and never going away.
This feeling isn’t something I can …