A Survey on Deep Generative 3D-aware Image Synthesis
arXiv:2210.14267 · doi:10.1145/3626193
Abstract
Recent years have seen remarkable progress in deep learning powered visual content creation. This includes deep generative 3D-aware image synthesis, which produces high-idelity images in a 3D-consistent manner while simultaneously capturing compact surfaces of objects from pure image collections without the need for any 3D supervision, thus bridging the gap between 2D imagery and 3D reality. The ield of computer vision has been recently captivated by the task of deep generative 3D-aware image synthesis, with hundreds of papers appearing in top-tier journals and conferences over the past few years (mainly the past two years), but there lacks a comprehensive survey of this remarkable and swift progress. Our survey aims to introduce new researchers to this topic, provide a useful reference for related works, and stimulate future research directions through our discussion section. Apart from the presented papers, we aim to constantly update the latest relevant papers along with corresponding implementations at https://weihaox.github.io/3D-aware-Gen.
Accepted to ACM Computing Surveys. Project page: https://weihaox.github.io/3D-aware-Gen
References in corpus (18)
- Instant Neural Graphics Primitives with a Multiresolution Hash Encoding
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
- NeRF++: Analyzing and Improving Neural Radiance Fields
- NeRF--: Neural Radiance Fields Without Known Camera Parameters
- Dual Octree Graph Networks for Learning Adaptive Volumetric Shape Representations
- CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis
- Novel View Synthesis with Diffusion Models
- StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generation
- Generative Neural Articulated Radiance Fields
- DragonDiffusion: Enabling Drag-style Manipulation on Diffusion Models
- Towards Open-World Text-Guided Face Image Generation and Manipulation
- Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering
- DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing
- The Stable Artist: Steering Semantics in Diffusion Latent Space
- Generative Novel View Synthesis with 3D-Aware Diffusion Models
- 3DShape2VecSet: A 3D Shape Representation for Neural Fields and Generative Diffusion Models
- VQ3D: Learning a 3D-Aware Generative Model on ImageNet
- Generating Images with 3D Annotations Using Diffusion Models