2306.09109] NAVI: Category-Agnostic Image Collections with High-Quality 3D Shape and Pose Annotations Skip to main

1905.10711] DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction Skip to main

2404.12385] MeshLRM: Large Reconstruction Model for High-Quality Meshes Happy Open Access Week from arXiv! YOU make open

2311.17261] SceneTex: High-Quality Texture Synthesis for Indoor Scenes via Diffusion Priors Skip to main content

2306.09109] NAVI: Category-Agnostic Image Collections with High-Quality 3D Shape and Pose Annotations Skip to main

2311.17261] SceneTex: High-Quality Texture Synthesis for Indoor Scenes via Diffusion Priors Skip to main content

2306.09109] NAVI: Category-Agnostic Image Collections with High-Quality 3D Shape and Pose Annotations Skip to main

2409.12957] 3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion The Scheduled Database Maintenance 2025-09

2409.12957] 3DTopia-XL: Scaling High-quality 3D Asset Generation via Primitive Diffusion Skip to main content We gratefully

2310.01406] HumanNorm: Learning Normal Diffusion Model for High-quality and Realistic 3D Human Generation Skip to

2411.07126] Edify Image: High-Quality Image Generation with Pixel Space Laplacian Diffusion Models Happy Open Access Week from arXiv! YOU

We introduce ROBIN: an MLM instruction-tuned with densely annotated relationships capable of constructing high-quality dense scene graphs

latent texture. We thoroughly validate TexFusion and show that we can efficiently generate diverse, high quality and globally

We introduce ROBIN: an MLM instruction-tuned with densely annotated relationships capable of constructing high-quality dense scene graphs

flash, in low-light environments. Our goal is to produce a high-quality rendering of

Song and 2 other authors View PDF Abstract: Denoising diffusion probabilistic models (DDPMs) have achieved high quality image generation without

flash, in low-light environments. Our goal is to produce a high-quality rendering of

experimental) Abstract: We propose RelitLRM, a Large Reconstruction Model (LRM) for generating high-quality Gaussian splatting representations

that this representation can be reconstructed using only multi-view image supervision and generates high-quality rendering results. More

distillation and consistency distillation to achieve few-step video generation, maintaining both high quality and diversity

for diffusion, and multi-stage training that enable us to directly generate high quality and high

7 other authors View PDF HTML (experimental) Abstract: Advances in 3D reconstruction have enabled high-quality 3D capture, but

for sampling synthetic pattern analogies, enables the creation of a large, high-quality synthetic training dataset

problems to be addressed in the future. Furthermore, we use this high-quality dataset to

the surfaces of 3D shapes, with the goal of synthesizing high-quality textures. Our approach

limitation, we propose consistency models, a new family of models that generate high quality samples by directly

distillation and consistency distillation to achieve few-step video generation, maintaining both high quality and diversity

for sampling synthetic pattern analogies, enables the creation of a large, high-quality synthetic training dataset

information into 3D space. We design a data generation pipeline to ensure high-quality 3D guidance throughout

to-image diffusion models have made significant advances in generating and editing high-quality images. As a

that calibrating the text conditioning of the source distribution can produce high-quality generation and

suite, achieving higher success rates and reduced task completion times. Moreover, the high-quality teleoperation demonstrations improve

the internal patch distribution from a single 3D textured shape and generates high-quality variations with fine

time. In this paper, we propose an efficient yet effective framework, HumanGaussian, that generates high-quality 3D humans with

simple yet non-trivial techniques that contribute to fast convergence speed and high-quality output. First, we

Text-to-image diffusion models have the remarkable ability to generate high-quality images with diverse

is able to use the same perspective warping to render high-quality images on two

the strengths of both approaches, providing a promising direction for high-quality visual generation with

text-to-image synthesis models have achieved an exceptional level of photorealism, generating high-quality images from arbitrary

excel at rendering photorealistic novel views of complex scenes. However, recovering a high-quality NeRF typically requires

that calibrating the text conditioning of the source distribution can produce high-quality generation and

Text-to-image diffusion models have the remarkable ability to generate high-quality images with diverse

is able to use the same perspective warping to render high-quality images on two

time. In this paper, we propose an efficient yet effective framework, HumanGaussian, that generates high-quality 3D humans with

investigate the use of Neural Radiance Fields (NeRF) to learn high quality 3D object category

PDF Abstract: We propose GS-LRM, a scalable large reconstruction model that can predict high-quality 3D Gaussian primitives

a single image of any object and can synthesize an accurate, high-quality relit image under

performance in dynamic environments. At the same time, to obtain high-quality tracking demonstrations, we

LRM-Zero, a Large Reconstruction Model (LRM) trained entirely on synthesized 3D data, achieving high-quality sparse-view 3D

normal training data. Instead of relying on large-scale annotated video datasets, we demonstrate high-quality video buffer estimation

train existing 2D diffusion models on these representations to generate 3D neural fields with high quality and diversity

the real world, we propose OmniObject3D, a large vocabulary 3D object dataset with massive high-quality real-scanned 3D

360$^{\circ}$ images. MDPs are more compact than previous 3D scene representations and enable high-quality, efficient new view

ambiguous regions. Our optimization shares information between two poses, which allows us to produce high fidelity and realistic

have created a 3D GAN that is both efficient and of high quality, while allowing for

View PDF HTML (experimental) Abstract: 3D Gaussian Splatting (3DGS) has recently revolutionized radiance field reconstruction, achieving high quality novel view synthesis

1) 3D and 2D diffusion guidance to effectively learn a high-quality static 3D asset

authors View PDF HTML (experimental) Abstract: Research on video generation has recently made tremendous progress, enabling high-quality videos to

in the image. We present our results over diverse images and prompts, demonstrating high-quality synthesis and

Tianwei Yin and 6 other authors View PDF HTML (experimental) Abstract: Diffusion models generate high-quality images but require

in this work, we develop a completely automated process to generate high-quality topic pages for

authors View PDF HTML (experimental) Abstract: Research on video generation has recently made tremendous progress, enabling high-quality videos to

of obtaining ground truth annotations. We present a system for mining high-quality 4D reconstructions from

the problem of the shortage of large repositories of high-quality 3D training models

We introduce a novel learning-based method to reconstruct the high-quality geometry and

challenging effects like specularities, shadows and occlusions. This allows us to perform high-quality view synthesis and

We introduce a novel learning-based method to reconstruct the high-quality geometry and

to-image diffusion models have made significant advances in generating and editing high-quality images. As a

challenging effects like specularities, shadows and occlusions. This allows us to perform high-quality view synthesis and

data. In this paper, we propose Instant3D, a novel method that generates high-quality and diverse

NeRF-based representations. However, a naive application of Gaussian splatting cannot generate high-quality animatable avatars and

that visual occlusions are implicitly taken into account. Extensive experiments demonstrate that our method can generate high-quality and realistic

the-shelf image diffusion models such as Stable Diffusion. Zero123++ excels in producing high-quality, consistent multi-view

object parts is required for generating diverse reconstructions with sharp textures. We also show high-quality text-to

We incorporate an RGBD panorama diffusion model to mitigate it, resulting in high-quality geometry. Extensive evaluation

scheme bypasses the need for view-inconsistent upsamplers and achieves high-quality multi-view consistent

that visual occlusions are implicitly taken into account. Extensive experiments demonstrate that our method can generate high-quality and realistic

the-shelf image diffusion models such as Stable Diffusion. Zero123++ excels in producing high-quality, consistent multi-view

NeRF-based representations. However, a naive application of Gaussian splatting cannot generate high-quality animatable avatars and

investigate the use of Neural Radiance Fields (NeRF) to learn high quality 3D object category

excel at rendering photorealistic novel views of complex scenes. However, recovering a high-quality NeRF typically requires

of 3D models but lack photorealistic appearance. We present an approach to automatically assign high-quality, realistic appearance models

have created a 3D GAN that is both efficient and of high quality, while allowing for

object parts is required for generating diverse reconstructions with sharp textures. We also show high-quality text-to

in this work, we develop a completely automated process to generate high-quality topic pages for

Tianwei Yin and 6 other authors View PDF HTML (experimental) Abstract: Diffusion models generate high-quality images but require

Somepalli and 4 other authors View PDF Abstract: Cutting-edge diffusion models produce images with high quality and customizability

data. In this paper, we propose Instant3D, a novel method that generates high-quality and diverse

synthesis and editing tasks. Recent advances in this field have also enabled high-quality 3D or video

Guidance, by Yongwei Chen and 5 other authors View PDF HTML (experimental) Abstract: Generating high-quality 3D assets from

advances in neural rendering, due to the scarcity of high-quality 3D datasets and

R. Chan and 10 other authors View PDF Abstract: Unsupervised generation of high-quality multi-view-consistent

the focus on dual-camera super-resolution (DCSR), which utilizes reference images for high-quality and high

and the initial diffusion noise map, which we utilize to consistently generate high-quality chrome balls. We

Comprehensive qualitative and quantitative evaluations demonstrate that HOI-Swap significantly outperforms existing methods, delivering high-quality video edits with

a neural inverse rendering pipeline called IRON that operates on photometric images and outputs high-quality 3D content in

on text-to-image synthesis (T2I) tasks. Despite their ability to generate high-quality yet creative images

our method on numerous inputs from various domains, showcasing a plethora of high quality complex semantic image