Stylizing 3D Scene via Implicit Representation and HyperNetwork

In this work, we aim to address the 3D scene stylization problem - generating stylized images of the scene at arbitrary novel view angles. A straightforward solution is to combine existing novel view synthesis and image/video style transfer approaches, which often leads to blurry results or inconsis...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings / IEEE Workshop on Applications of Computer Vision S. 215 - 224
Hauptverfasser: Chiang, Pei-Ze, Tsai, Meng-Shiun, Tseng, Hung-Yu, Lai, Wei-Sheng, Chiu, Wei-Chen
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.01.2022
Schlagworte:
3D Computer Vision Neural rendering
Computer vision
Optimization
Prediction algorithms
Predictive models
Solid modeling
Three-dimensional displays
Training
ISSN:2642-9381
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this work, we aim to address the 3D scene stylization problem - generating stylized images of the scene at arbitrary novel view angles. A straightforward solution is to combine existing novel view synthesis and image/video style transfer approaches, which often leads to blurry results or inconsistent appearance. Inspired by the high-quality results of the neural radiance fields (NeRF) method, we propose a joint framework to directly render novel views with the desired style. Our framework consists of two components: an implicit representation of the 3D scene with the neural radiance fields model, and a hypernetwork to transfer the style information into the scene representation. To alleviate the training difficulties and memory burden, we propose a two-stage training procedure and a patch sub-sampling approach to optimize the style and content losses with the neural radiance fields model. After optimization, our model is able to render consistent novel views at arbitrary view angles with arbitrary style. Both quantitative evaluation and human subject study have demonstrated that the proposed method generates faithful stylization results with consistent appearance across different views.
ISSN:2642-9381
DOI:10.1109/WACV51458.2022.00029