CMUが富士通の研究者と共同で、動的3次元構造表現のブレークスルーを達成(CMU Collaborates with Fujitsu Researchers on Breakthrough in Dynamic 3D Structure Representation)


2023-06-14 カーネギーメロン大学

Dynamic Light Field Network (DyLiN) accommodates dynamic scene deformations, such as in avatar animation, where facial expressions can be used as controllable input attributes.

◆カーネギーメロン大学と富士通の研究者は、2D画像を3D構造に変換する新しい手法を開発しました。この手法は、非剛体変形や位相変化を処理し、従来の静的光場ネットワークよりも優れた性能を持つDynamic Light Field Network(DyLiN)を利用しています。DyLiNは、入力光線から正規化光線への変形フィールドを学習し、それを高次元空間に取り扱いやすくします。


DyLiN: Making Light Field Networks Dynamic

Heng Yu, Joel Julin, Zoltan A. Milacski, Koichiro Niinuma, Laszlo A. Jeni
arXiv  Submitted on 24 Mar 2023

Light Field Networks, the re-formulations of radiance fields to oriented rays, are magnitudes faster than their coordinate network counterparts, and provide higher fidelity with respect to representing 3D structures from 2D observations. They would be well suited for generic scene representation and manipulation, but suffer from one problem: they are limited to holistic and static scenes. In this paper, we propose the Dynamic Light Field Network (DyLiN) method that can handle non-rigid deformations, including topological changes. We learn a deformation field from input rays to canonical rays, and lift them into a higher dimensional space to handle discontinuities. We further introduce CoDyLiN, which augments DyLiN with controllable attribute inputs. We train both models via knowledge distillation from pretrained dynamic radiance fields. We evaluated DyLiN using both synthetic and real world datasets that include various non-rigid deformations. DyLiN qualitatively outperformed and quantitatively matched state-of-the-art methods in terms of visual fidelity, while being 25 – 71x computationally faster. We also tested CoDyLiN on attribute annotated data and it surpassed its teacher model. Project page: this https URL .