3d-aware encoding for style-based neural radiance fields
YJ Li, T Xu, B Wu, N Zheng, X Dai, A Pumarola… - arXiv preprint arXiv …, 2022 - arxiv.org
We tackle the task of NeRF inversion for style-based neural radiance fields,(eg, StyleNeRF).
In the task, we aim to learn an inversion function to project an input image to the latent space …
In the task, we aim to learn an inversion function to project an input image to the latent space …
Self-supervised Learning for Enhancing Geometrical Modeling in 3D-Aware Generative Adversarial Network
J Guo, X Xu, H Zhao - arXiv preprint arXiv:2312.11856, 2023 - arxiv.org
3D-aware Generative Adversarial Networks (3D-GANs) currently exhibit artifacts in their 3D
geometrical modeling, such as mesh imperfections and holes. These shortcomings are …
geometrical modeling, such as mesh imperfections and holes. These shortcomings are …
Federated Multi-View Synthesizing for Metaverse
Y Guo, Z Qin, X Tao, GY Li - IEEE Journal on Selected Areas in …, 2023 - ieeexplore.ieee.org
The metaverse is expected to provide immersive entertainment, education, and business
applications. However, virtual reality (VR) transmission over wireless networks is data-and …
applications. However, virtual reality (VR) transmission over wireless networks is data-and …
CMC: Few-shot Novel View Synthesis via Cross-view Multiplane Consistency
H Zhu, Z Chen - 2024 IEEE Conference Virtual Reality and 3D …, 2024 - ieeexplore.ieee.org
Neural Radiance Field (NeRF) has shown impressive results in novel view synthesis,
particularly in Virtual Reality (VR) and Augmented Reality (AR), thanks to its ability to …
particularly in Virtual Reality (VR) and Augmented Reality (AR), thanks to its ability to …
Facelit: Neural 3d relightable faces
A Ranjan, KM Yi, JHR Chang… - Proceedings of the IEEE …, 2023 - openaccess.thecvf.com
We propose a generative framework, FaceLit, capable of generating a 3D face that can be
rendered at various user-defined lighting conditions and views, learned purely from 2D …
rendered at various user-defined lighting conditions and views, learned purely from 2D …
Survey on controlable image synthesis with deep learning
S Zhang, J Li, L Yang - arXiv preprint arXiv:2307.10275, 2023 - arxiv.org
Image synthesis has attracted emerging research interests in academic and industry
communities. Deep learning technologies especially the generative models greatly inspired …
communities. Deep learning technologies especially the generative models greatly inspired …
SinMPI: Novel View Synthesis from a Single Image with Expanded Multiplane Images
G Pu, PS Wang, Z Lian - SIGGRAPH Asia 2023 Conference Papers, 2023 - dl.acm.org
Single-image novel view synthesis is a challenging and ongoing problem that aims to
generate an infinite number of consistent views from a single input image. Although …
generate an infinite number of consistent views from a single input image. Although …
Nerffacelighting: Implicit and disentangled face lighting representation leveraging generative prior in neural radiance fields
K Jiang, SY Chen, H Fu, L Gao - ACM Transactions on Graphics, 2023 - dl.acm.org
3D-aware portrait lighting control is an emerging and promising domain, thanks to the recent
advance of generative adversarial networks and neural radiance fields. Existing solutions …
advance of generative adversarial networks and neural radiance fields. Existing solutions …
CraftsMan: High-fidelity Mesh Generation with 3D Native Generation and Interactive Geometry Refiner
W Li, J Liu, R Chen, Y Liang, X Chen, P Tan… - arXiv preprint arXiv …, 2024 - arxiv.org
We present a novel generative 3D modeling system, coined CraftsMan, which can generate
high-fidelity 3D geometries with highly varied shapes, regular mesh topologies, and detailed …
high-fidelity 3D geometries with highly varied shapes, regular mesh topologies, and detailed …
Adversarial Generation of Hierarchical Gaussians for 3D Generative Model
S Hyun, JP Heo - arXiv preprint arXiv:2406.02968, 2024 - arxiv.org
Most advances in 3D Generative Adversarial Networks (3D GANs) largely depend on ray
casting-based volume rendering, which incurs demanding rendering costs. One promising …
casting-based volume rendering, which incurs demanding rendering costs. One promising …