image-generation

3D-Aware Image Synthesis

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in image-generation

Trend

Dataset

Best Model

Actions

FFHQ 256 x 256

CelebAMask-HQ

FFHQ 512 x 512 - 4x upscaling

Libraries

i

Use these libraries to find image-generation models and implementations

Datasets

FFHQ

CelebAMask-HQ

SynthRAD2023

3D-Point Cloud dataset of various geometrical terrains

Subtasks

No subtasks available.

Most implemented papers

pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis

Jiajun Wu, Gordon Wetzstein, Petr Kellnhofer, Eric Chan, M. Monteiro•Tue Dec 01 2020

We have witnessed rapid progress on 3D-aware image synthesis, leveraging recent advances in generative visual models and neural rendering. Existing approaches how-ever fall short in two ways: first, they may lack an under-lying 3D representation or rely on view-inconsistent rendering, hence synthesizing images that are not multi-view consistent; second, they often depend upon representation network architectures that are not expressive enough, and their results thus lack in image quality. We propose a novel generative model, named Periodic Implicit Generative Adversarial Networks (π-GAN or pi-GAN), for high-quality 3D-aware image synthesis. π-GAN leverages neural representations with periodic activation functions and volumetric rendering to represent scenes as view-consistent radiance fields. The proposed approach obtains state-of-the-art results for 3D-aware image synthesis with multiple real and synthetic datasets.

Content

FFHQ 512 x 512 - 4x upscaling

937 0

Paper Graph

Pix2NeRF: Unsupervised Conditional $\pi$-GAN for Single Image to Neural Radiance Fields Translation

L. Gool, Anton Obukhov, Dengxin Dai, Shengqu Cai•Fri Feb 25 2022

A pipeline to generate Neural Radiance Fields (NeRF) of an object or a scene of a specific class, conditioned on a single input image, based on $\pi$-GAN, a generative model for unconditional 3D-aware image synthesis, which maps random latent codes to radiance fields of a class of objects.

109 0

Paper Graph

GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis

Andreas Geiger, Michael Niemeyer, Yiyi Liao, Katja Schwarz•Sat Jul 04 2020

This paper proposes a generative model for radiance fields which have recently proven successful for novel view synthesis of a single scene, and introduces a multi-scale patch-based discriminator to demonstrate synthesis of high-resolution images while training the model from unposed 2D images alone.

947 0

Paper Graph

CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesis

Bingbing Ni, Lingxi Xie, Peng Zhou, Qi Tian•Mon Oct 18 2021

CIPS-3D is presented, a style-based, 3D-aware generator that is composed of a shallow NeRF network and a deep implicit neural representation (INR) network that synthesizes each pixel value independently without any spatial convolution or upsampling operation.

178 0

Paper Graph

A Shading-Guided Generative Implicit Model for Shape-Accurate 3D-Aware Image Synthesis

Chen Change Loy, C. Theobalt, Xingang Pan, Bo Dai, Xudong Xu•Thu Oct 28 2021

This work proposes a novel shading-guided generative implicit model that is able to learn a starkly improved shape representation and demonstrates improved performance of the approach on 3D shape reconstruction against existing methods, and its applicability on image relighting.

101 0

Paper Graph

FENeRF: Face Editing in Neural Radiance Fields

Yebin Liu, Qi Zhang, Xuan Wang, Jue Wang, Yong Zhang, Jingxiang Sun, Xiaoyu Li•Mon Nov 29 2021

This work proposes FENeRF, a 3D-aware generator that can produce view-consistent and locally-editable portrait images and reveals that Joint learning semantics and texture helps to generate finer geometry.

164 0

Paper Graph

Generative Occupancy Fields for 3D Surface-Aware Image Synthesis

Dahua Lin, Xingang Pan, Bo Dai, Xudong Xu•Sun Oct 31 2021

GOF is proposed, a novel model based on generative radiance fields that can synthesize high-quality images with 3D consistency and simultaneously learn compact and smooth object surfaces that combines the merits of two representations in a unified framework.

63 0

Paper Graph

3D-Aware Semantic-Guided Generative Model for Human Synthesis

N. Sebe, Aliaksandr Siarohin, H. Tang, E. Sangineto, Wei Wang, Zhun Zhong, Jichao Zhang•Wed Dec 01 2021

A 3D-aware Semantic-Guided Generative Model (3D-SGAN) for human image synthesis, which combines a GNeRF with a texture generator that can learn 3D human representations with a photo-realistic, controllable generation.

38 0

Paper Graph

3D-aware Image Synthesis via Learning Structural and Textural Representations

Bolei Zhou, Ceyuan Yang, Sida Peng, Yinghao Xu, Yujun Shen•Sun Dec 19 2021

This work proposes a novel framework, termed as VolumeGAN, for high-fidelity 3D-aware image synthesis, through explicitly learning a structural representation and a textural representation in a Generative Adversarial Network.

139 0

Paper Graph

XraySyn: Realistic View Synthesis From a Single Radiograph Through CT Priors

Jiebo Luo, R. Chellappa, Cheng Peng, Haofu Liao, G. Wong, S. Zhou•Thu Dec 03 2020

By gaining an understanding of radiography in 3D space, the method can be applied to radiograph bone extraction and suppression without requiring groundtruth bone labels and is the first work on radiograph view synthesis.

14 0

Paper Graph

Adding a benchmark result helps the community track progress.