Efficient Geometry-aware 3D Generative Adversarial Networks
01 Jun 2022
TL;DR: In this article , a hybrid explicit implicit network architecture is proposed to generate high-resolution multi-view-consistent images in real-time but also produce high-quality 3D geometry.
read more
Abstract: Unsupervised generation of high-quality multi-view-consistent images and 3D shapes using only collections of single-view 2D photographs has been a long-standing challenge. Existing 3D GANs are either compute intensive or make approximations that are not 3D-consistent; the former limits quality and resolution of the generated images and the latter adversely affects multi-view consistency and shape quality. In this work, we improve the computational efficiency and image quality of 3D GANs without overly relying on these approximations. We introduce an expressive hybrid explicit implicit network architecture that, together with other design choices, synthesizes not only high-resolution multi-view-consistent images in real time but also produces high-quality 3D geometry. By decoupling feature generation and neural rendering, our framework is able to leverage state-of-the-art 2D CNN generators, such as StyleGAN2, and inherit their efficiency and expressiveness. We demonstrate state-of-the-art 3D-aware synthesis with FFHQ and AFHQ Cats, among other experiments.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
TensoRF: Tensorial Radiance Fields
TL;DR: Zhang et al. as discussed by the authors modeled the radiance field of a scene as a 4D tensor, which represents a 3D voxel grid with per-voxel multi-channel features.
Magic3D: High-Resolution Text-to-3D Content Creation
Chen-Hsuan Lin,Jun Gao,Luming Tang,Towaki Takikawa,Xiaohui Zeng,Xun Huang,Karsten Kreis,Sanja Fidler,Mingyu Li,Tsung-Yi Lin +9 more
- 01 Jun 2023
TL;DR: Magic3D is a novel text-to-3D content creation method that significantly accelerates and improves the quality of 3D model generation compared to DreamFusion.
348
Data augmentation: A comprehensive survey of modern approaches
TL;DR: Data augmentation is the most effective way of alleviating the problem of data collection and annotation processes and consumes a lot of time and resources as mentioned in this paper , which is the main goal of data augmentation, to increase the volume, quality and diversity of training data.
318
RealFusion 360° Reconstruction of Any Object from a Single Image
Luke Melas-Kyriazi,Iro Laina,Christian Rupprecht,Andrea Vedaldi +3 more
- 21 Feb 2023
TL;DR: This work takes an off-the-self conditional image generator based on diffusion and engineer a prompt that encourages it to “dream up” novel views of the object, and fuse the given input view, the conditional prior, and other regularizers into a final, consistent reconstruction.
248
K-Planes: Explicit Radiance Fields in Space, Time, and Appearance
Sara Fridovich-Keil,Giacomo Meanti,Frederik Warburg,Benjamin Recht,Angjoo Kanazawa +4 more
- 01 Jun 2023
TL;DR: K-planes is a novel white-box radiance field model that efficiently represents high-dimensional scenes. It uses planes to factorize scenes, enabling easy addition of dimension-specific priors and natural decomposition of static and dynamic components. K-planes achieves competitive reconstruction fidelity with low memory usage and fast optimization.
179
References
A Style-Based Generator Architecture for Generative Adversarial Networks
Tero Karras,Samuli Laine,Timo Aila +2 more
- 15 Jun 2019
TL;DR: This paper proposed an alternative generator architecture for GANs, borrowing from style transfer literature, which leads to an automatically learned, unsupervised separation of high-level attributes (e.g., pose and identity when trained on human faces) and stochastic variation in the generated images.
•Posted Content
GANs Trained by a Two Time-Scale Update Rule Converge to a Nash Equilibrium
Martin Heusel,Hubert Ramsauer,Thomas Unterthiner,Bernhard Nessler,Günter Klambauer,Sepp Hochreiter +5 more
TL;DR: In this article, a two time-scale update rule (TTUR) was proposed for training GANs with stochastic gradient descent on arbitrary GAN loss functions, which has an individual learning rate for both the discriminator and the generator.
9.2K
ArcFace: Additive Angular Margin Loss for Deep Face Recognition
Jiankang Deng,Jia Guo,Niannan Xue,Stefanos Zafeiriou +3 more
- 15 Jun 2019
TL;DR: This paper presents arguably the most extensive experimental evaluation against all recent state-of-the-art face recognition methods on ten face recognition benchmarks, and shows that ArcFace consistently outperforms the state of the art and can be easily implemented with negligible computational overhead.
•Proceedings Article
Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
Alec Radford,Luke Metz,Soumith Chintala +2 more
- 01 Jan 2016
TL;DR: Deep convolutional generative adversarial networks (DCGANs) as discussed by the authors learn a hierarchy of representations from object parts to scenes in both the generator and discriminator for unsupervised learning.
7.3K