paper-with-me

Papers

CtrlNeRF: The Generative Neural Radiation Fields for the Controllable Synthesis of High-fidelity 3D-Aware Images

2024-12-01 · Jian Liu, Zhen Yu

The neural radiance field (NERF) advocates learning the continuous representation of 3D geometry through a multilayer perceptron (MLP). By integrating this into a generative model, the generative neural radiance field (GRAF) is capable of producing images from random noise z without 3D supervision. In practice, the shape and appearance are modeled by z_s and z_a, respectively, to manipulate them separately during inference. However, it is challenging to represent multiple scenes using a solitary MLP and precisely control the generation of 3D geometry in terms of shape and appearance. In this paper, we introduce a controllable generative model (i.e. \textbf{CtrlNeRF}) that uses a single MLP network to represent multiple scenes with shared weights. Consequently, we manipulated the shape and appearance codes to realize the controllable generation of high-fidelity images with 3D consistency. Moreover, the model enables the synthesis of novel views that do not exist in the training sets via camera pose alteration and feature interpolation. Extensive experiments were conducted to demonstrate its superiority in 3D-aware image generation compared to its counterparts.

📄 PDF Abstract BibTeX arXiv:2412.00754

Code (0)

등록된 구현이 없습니다.

Tasks

3D geometryImage GenerationNeRF

Similar Papers 제목 키워드 기반

CGOF++: Controllable 3D Face Synthesis with Conditional Generative Occupancy Fields

2022-11-23 · Keqiang Sun, Shangzhe Wu, Ning Zhang, Zhaoyang Huang 외

Capitalizing on the recent advances in image generation models, existing controllable face image synthesis methods are able to generate high-fidelity images with some levels of controllability, e.g., controlling the shap…

Face GenerationImage GenerationNeRF

GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields

2020-11-24 · CVPR 2021 1 · Michael Niemeyer, Andreas Geiger

Deep generative models allow for photorealistic image synthesis at high resolutions. But for many applications, this is not enough: content creation also needs to be controllable. While several recent works investigate h…

Image GenerationNeural Rendering

Controllable 3D Face Synthesis with Conditional Generative Occupancy Fields

2022-06-16 · Keqiang Sun, Shangzhe Wu, Zhaoyang Huang, Ning Zhang 외

Capitalizing on the recent advances in image generation models, existing controllable face image synthesis methods are able to generate high-fidelity images with some levels of controllability, e.g., controlling the shap…

Face GenerationImage GenerationNeRF

DisCoScene: Spatially Disentangled Generative Radiance Fields for Controllable 3D-aware Scene Synthesis

2022-12-22 · CVPR 2023 1 · Yinghao Xu, Menglei Chai, Zifan Shi, Sida Peng 외

Existing 3D-aware image synthesis approaches mainly focus on generating a single canonical object and show limited capacity in composing a complex scene containing a variety of objects. This work presents DisCoScene: a 3…

3D-Aware Image SynthesisImage GenerationObject

VeRi3D: Generative Vertex-based Radiance Fields for 3D Controllable Human Image Synthesis

2023-09-09 · ICCV 2023 1 · Xinya Chen, Jiaxin Huang, Yanrui Bin, Lu Yu 외

Unsupervised learning of 3D-aware generative adversarial networks has lately made much progress. Some recent work demonstrates promising results of learning human generative models using neural articulated radiance field…

Image Generation