Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Gaussian Shell Maps are a new neural scene representation that connects fields and 3D Gaussians. This representation unlocks the full potential of 3D Gaussian splatting for generative AI applications, such as 3D avatar generation. 1/2

52,449 görüntüleme • 2 yıl önce •via X (Twitter)

3 Yorum

Gordon Wetzstein profil fotoğrafı
Gordon Wetzstein2 yıl önce

With @AbdalRameen, @toomanyyifans, @Vivianszf1, @YinghaoXu1, @Po_lhr, @zfkuang1, @CQFHK, Dit-Yan Yeung 2/2

Hermes ᯅ profil fotoğrafı
Hermes ᯅ2 yıl önce

This looks really interesting. Over the last few weeks I’ve seen papers come out on how to record motion but this looks like it can synthesize the motion which is a whole new level.

Wieland Morgenstern profil fotoğrafı
Wieland Morgenstern2 yıl önce

Lovely looking results, kudos! Bit unclear to me how the shells are actually formed, and how the Gaussians are placed in them. I put my notes from reading the paper here, including the stuff that I don't understand:

Benzer Videolar

Nvidia announces GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning paper page: Gaussian splatting has emerged as a powerful 3D representation that harnesses the advantages of both explicit (mesh) and implicit (NeRF) 3D representations. In this paper, we seek to leverage Gaussian splatting to generate realistic animatable avatars from textual descriptions, addressing the limitations (e.g., flexibility and efficiency) imposed by mesh or NeRF-based representations. However, a naive application of Gaussian splatting cannot generate high-quality animatable avatars and suffers from learning instability; it also cannot capture fine avatar geometries and often leads to degenerate body parts. To tackle these problems, we first propose a primitive-based 3D Gaussian representation where Gaussians are defined inside pose-driven primitives to facilitate animation. Second, to stabilize and amortize the learning of millions of Gaussians, we propose to use neural implicit fields to predict the Gaussian attributes (e.g., colors). Finally, to capture fine avatar geometries and extract detailed meshes, we propose a novel SDF-based implicit mesh learning approach for 3D Gaussians that regularizes the underlying geometries and extracts highly detailed textured meshes. Our proposed method, GAvatar, enables the large-scale generation of diverse animatable avatars using only text prompts. GAvatar significantly surpasses existing methods in terms of both appearance and geometry quality, and achieves extremely fast rendering (100 fps) at 1K resolution.

AK

140,960 görüntüleme • 2 yıl önce