Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

35,456 Aufrufe • vor 1 Tag •via X (Twitter)

11 Kommentare

Profilbild von cognito
cognitovor 1 Tag

You can run the same experiment using colmap + gaussian splat and get arguably better results

Profilbild von neilson
neilsonvor 1 Tag

some reasons this is slightly different 1. these rooms aren't real 2. any input view can be changed in realtime to change the entire scene

Profilbild von Marcin Kobylecki
Marcin Kobyleckivor 1 Tag

I gave it a try, and I must admit I actually had quite a bit of fun with it. There is one issue, however: it lacks the production quality required for a film; the realism comes across better than the 3D animation style.

Profilbild von EX3D
EX3Dvor 1 Tag

Fucking incredible.

Profilbild von Gio Gargiulo
Gio Gargiulovor 23 Stunden

What am I looking at?

Profilbild von professional biaser
professional biaservor 1 Tag

absolutely mental. literally alien tech.

Profilbild von Rowdy Burns
Rowdy Burnsvor 1 Tag

What software?

Profilbild von tommy
tommyvor 15 Stunden

fuck…

Profilbild von Phil Gara
Phil Garavor 1 Tag

Amazing!

Profilbild von metamike
metamikevor 1 Tag

Tf is this

Profilbild von Broadstreet
Broadstreetvor 21 Stunden

Excellent

Ähnliche Videos

MaterialFusion Enhancing Inverse Rendering with Material Diffusion Priors discuss: Recent works in inverse rendering have shown promise in using multi-view images of an object to recover shape, albedo, and materials. However, the recovered components often fail to render accurately under new lighting conditions due to the intrinsic challenge of disentangling albedo and material properties from input images. To address this challenge, we introduce MaterialFusion, an enhanced conventional 3D inverse rendering pipeline that incorporates a 2D prior on texture and material properties. We present StableMaterial, a 2D diffusion model prior that refines multi-lit data to estimate the most likely albedo and material from given input appearances. This model is trained on albedo, material, and relit image data derived from a curated dataset of approximately ~12K artist-designed synthetic Blender objects called BlenderVault. we incorporate this diffusion prior with an inverse rendering framework where we use score distillation sampling (SDS) to guide the optimization of the albedo and materials, improving relighting performance in comparison with previous work. We validate MaterialFusion's relighting performance on 4 datasets of synthetic and real objects under diverse illumination conditions, showing our diffusion-aided approach significantly improves the appearance of reconstructed objects under novel lighting conditions. We intend to publicly release our BlenderVault dataset to support further research in this field.

AK

22,959 Aufrufe • vor 2 Jahren