V
主页
GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting
发布人
【加群】 一起来刷arxiv,请加vx: pwbot02(请备注:b站arxiv) 【彩蛋】 可以试试/ask + 你的提问和本篇论文进行交流 【论文标题】 GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting 【论文简述】 本论文介绍了一种名为高斯编辑器(GaussianEditor)的创新而高效的三维编辑算法。传统的三维编辑方法常常无法真实地描绘复杂场景,因其依赖于网格和点云等表示方式。而基于隐式三维表示方法,如神经辐射场(Neural Radiance Field, NeRF),可以有效地渲染复杂场景,但处理速度较慢且对特定场景区域的控制有限。为了应对这些挑战,本文提出了一种基于高斯飞溅(Gaussian Splatting)的创新和高效的三维编辑算法,名为高斯编辑器。高斯编辑器通过我们提出的高斯语义追踪(Gaussian semantic tracing)技术,增强了编辑的精度和控制性。此外,我们还提出了分层高斯飞溅(Hierarchical Gaussian splatting, HGS)算法,在二维扩散模型的随机生成指导下实现了稳定且精细的结果。我们还针对对象的高效移除和融合开发了编辑策略,这是现有方法中具有挑战性的任务。通过全面的实验证明,高斯编辑器在控制性、效果和性能快速性方面表现优异,标志着三维编辑领域的重要进展。 【引导阅读的问题】 如何通过高斯编辑器有效地实现对三维场景的精确编辑和控制? 【论文链接】 https://arxiv.org/pdf/2311.14521
打开封面
下载高清视频
观看高清视频
视频下载器
LucidDreamer: Domain-free Generation of 3D Gaussian Splatting Scenes
Editing Personality for LLMs
高斯涂抹文本生成三维模型
GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian Spla
Progressive3D: Progressively Local Editing for Text-to-3D Content Creation with
Controllable Music Production with Diffusion Models and Guidance Gradients
Drivable 3D Gaussian Avatars
Localizing and Editing Knowledge in Text-to-Image Generative Models
LLM Lies: Hallucinations are not Bugs, but Features as Adversarial Examples
AutoStory: Generating Diverse Storytelling Images with Minimal Human Effort
Contrastive Chain-of-Thought Prompting
TEAL: Tokenize and Embed ALL for Multi-modal Large Language Models
ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAs
SyncDreamer:基于单图像生成3D物体新方法
PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics
PALP: Prompt Aligned Personalization of Text-to-Image Models
FreeControl:实现任意文本到图像扩散模型的无训练空间控制
An Image is Worth Multiple Words: Learning Object Level Concepts using Multi-Con
Kosmos-G: Generating Images in Context with Multimodal Large Language Models
More Agents Is All You Need
In-Context Learning Creates Task Vectors
多模态协同学习模型
一种基于全局优化的实时SLAM算法
GraphLLM: Boosting Graph Reasoning Ability of Large Language Model
Wonder3D: Single Image to 3D using Cross-Domain Diffusion
Visual In-Context Prompting
OceanGPT: A Large Language Model for Ocean Science Tasks
The Chosen One: Consistent Characters in Text-to-Image Diffusion Models
SHINOBI: Shape and Illumination using Neural Object Decomposition via BRDF Optim
基于高斯混合掩码的视觉变换器在小数据集上的提升
多模态模型的规模化:预训练和指导调整
GARField: Group Anything with Radiance Fields
Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection
Can a student Large Language Model perform as well as it's teacher?
Prompt2Model实现提示驱动的AutoML
Conditional Diffusion Distillation
图神经网络中的关键子图发现:不变性图变换器
VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Mode
Implicit Diffusion: Efficient Optimization through Stochastic Sampling
EvoPrompt: AI Prompt Optimizer