High-Quality Rendering Dataset of Objav
type="video/mp4">
- Samples of the HQR-Objaverse. From top to bottom, there are the RGB, Albedo, Normal and Depth images.
+ Samples of the dataset. From top to bottom, there are the RGB, Albedo, Normal and Depth images.
Introduction
- High-Quality Rendering Dataset of Objaverse (HQR-Objaverse) is rendered using the TIDE renderer on Objaverse with A10 for about 2000 GPU hours, yielding 3000,0000 number of albedo, RGB, Depth, and Normal images. We proposed a rendering framework for high quality and high speed dataset rendering. The framework is a hybrid of rasterization and path tracing, the first ray-scene intersection is obtained by hardware rasterization and accurate indirect lighting by full hardware path tracing. Additionally, we using adaptive sampling, denoiser and path-guiding to further speed up the rendering time. In this rendering framework, we render 38 views of a centered object, including 24 views at elevation range from 5° to 30°, rotation = {r × 15° | r ∈ [0, 23]}, and 12 views at elevation from -5° to 5°, rotation = {r × 30° | r ∈ [0, 11]}, and 2 views for top and bottom respectively. In addition, we mannuly split the objaverse dataset into 10 general categories including Human-Shape, Animals, Daily-Used, Furnitures, Buildings&&Outdoor, Transportations, Plants, Food, Electronics and Poor-quality.
+ High-Quality Rendering Dataset of Objaverse (HQR-Objaverse) is rendered using the TIDE renderer on Objaverse with A10 for about 2000 GPU hours, yielding 30,000,000 images of albedo, RGB, Depth, and Normal map. We proposed a rendering framework for high quality and high speed dataset rendering. The framework is a hybrid of rasterization and path tracing, the first ray-scene intersection is obtained by hardware rasterization and accurate indirect lighting by full hardware path tracing. Additionally, we using adaptive sampling, denoiser and path-guiding to further speed up the rendering time. In this rendering framework, we render 38 views of a centered object, including 24 views at elevation range from 5° to 30°, rotation = {r × 15° | r ∈ [0, 23]}, and 12 views at elevation from -5° to 5°, rotation = {r × 30° | r ∈ [0, 11]}, and 2 views for top and bottom respectively. In addition, we mannuly split the objaverse dataset into 10 general categories including Human-Shape, Animals, Daily-Used, Furnitures, Buildings&&Outdoor, Transportations, Plants, Food, Electronics and Poor-quality.
@@ -210,10 +210,11 @@ Video
Application
- 
+
- We have used HQR-Objaverse for training MultiView Normal-Depth diffusion model (ND-MV) and depth-condition MultiView Albedo diffusion model (Albedo-MV), which are employed for 3D object generation through score-distillation sampling (SDS) in RichDreamer.
+ We have used HQR-Objaverse for training MultiView Normal-Depth diffusion model (ND-MV) and depth-condition MultiView Albedo diffusion model (Albedo-MV), which are employed for 3D object generation through score-distillation sampling (SDS) in RichDreamer .
+
- High-Quality Rendering Dataset of Objaverse (HQR-Objaverse) is rendered using the TIDE renderer on Objaverse with A10 for about 2000 GPU hours, yielding 3000,0000 number of albedo, RGB, Depth, and Normal images. We proposed a rendering framework for high quality and high speed dataset rendering. The framework is a hybrid of rasterization and path tracing, the first ray-scene intersection is obtained by hardware rasterization and accurate indirect lighting by full hardware path tracing. Additionally, we using adaptive sampling, denoiser and path-guiding to further speed up the rendering time. In this rendering framework, we render 38 views of a centered object, including 24 views at elevation range from 5° to 30°, rotation = {r × 15° | r ∈ [0, 23]}, and 12 views at elevation from -5° to 5°, rotation = {r × 30° | r ∈ [0, 11]}, and 2 views for top and bottom respectively. In addition, we mannuly split the objaverse dataset into 10 general categories including Human-Shape, Animals, Daily-Used, Furnitures, Buildings&&Outdoor, Transportations, Plants, Food, Electronics and Poor-quality. + High-Quality Rendering Dataset of Objaverse (HQR-Objaverse) is rendered using the TIDE renderer on Objaverse with A10 for about 2000 GPU hours, yielding 30,000,000 images of albedo, RGB, Depth, and Normal map. We proposed a rendering framework for high quality and high speed dataset rendering. The framework is a hybrid of rasterization and path tracing, the first ray-scene intersection is obtained by hardware rasterization and accurate indirect lighting by full hardware path tracing. Additionally, we using adaptive sampling, denoiser and path-guiding to further speed up the rendering time. In this rendering framework, we render 38 views of a centered object, including 24 views at elevation range from 5° to 30°, rotation = {r × 15° | r ∈ [0, 23]}, and 12 views at elevation from -5° to 5°, rotation = {r × 30° | r ∈ [0, 11]}, and 2 views for top and bottom respectively. In addition, we mannuly split the objaverse dataset into 10 general categories including Human-Shape, Animals, Daily-Used, Furnitures, Buildings&&Outdoor, Transportations, Plants, Food, Electronics and Poor-quality.
+
- We have used HQR-Objaverse for training MultiView Normal-Depth diffusion model (ND-MV) and depth-condition MultiView Albedo diffusion model (Albedo-MV), which are employed for 3D object generation through score-distillation sampling (SDS) in RichDreamer. + We have used HQR-Objaverse for training MultiView Normal-Depth diffusion model (ND-MV) and depth-condition MultiView Albedo diffusion model (Albedo-MV), which are employed for 3D object generation through score-distillation sampling (SDS) in RichDreamer .
+