GGRt: Pose-free Generalizable 3D Gaussian Splatting in Real-time
A novel approach, GGRt, enables pose-free 3D Gaussian splatting for real-time rendering, outperforming existing methods.
Consistent Image to 3D View Synthesis via Geometry-aware Diffusion Models
A novel framework, Consistent-1-to-3, addresses the challenge of maintaining 3D consistency in view synthesis through a two-stage generative model.
Ctrl123: Consistent Novel View Synthesis via Closed-Loop Transcription
Closed-loop transcription improves pose and appearance consistency in novel view synthesis.
GeoGaussian: Geometry-aware Gaussian Splatting for Scene Rendering
Preserving accurate geometry in Gaussian models enhances rendering quality.
Consistent Image to 3D View Synthesis via Geometry-aware Diffusion Models
Consistent-1-to-3 is a generative framework that significantly mitigates the issue of maintaining 3D consistency across different views in novel view synthesis.
GeoGaussian: Geometry-aware Gaussian Splatting for Scene Rendering
GeoGaussian enhances 3D Gaussian rendering quality by preserving accurate geometry and introducing novel densification strategies.
GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesis
A new approach, GPS-Gaussian, enables real-time novel view synthesis with high fidelity using pixel-wise Gaussian splatting.
GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesis
提案されたGPS-Gaussianは、稀疎ビューカメラ設定下でリアルタイムの高品質な人物の新しい視点合成を可能にする。
DreamComposer: Controllable 3D Object Generation via Multi-View Conditions
DreamComposer enhances existing view-aware diffusion models by injecting multi-view conditions, enabling controllable 3D object generation.
Improving Novel View Synthesis through Self-Augmentation with Synthesized Views
Leveraging a neural radiance field (NeRF) model's own view synthesis capability to generate synthetic views and use them to augment the training data, leading to improved novel view synthesis performance.