D3GS is proposed, a Depth-DINO-Diffusion guided sparse-view Gaussian reconstruction framework that jointly enhances geometry and appearance and injects generative priors into an iterative optimization strategy, enhancing high-frequency geometric and appearance details within the Gaussian representation.
Abstract
Novel view synthesis from sparse inputs remains challenging for 3D Gaussian Splatting (3DGS) due to ambiguous geometry, cross-view inconsistency, and missing details in under-constrained regions, resulting in degraded reconstruction and unstable rendering. To tackle these issues, we propose D$^{3}$GS, a Depth-DINO-Diffusion guided sparse-view Gaussian reconstruction framework that jointly enhances geometry and appearance. D$^{3}$GS first recovers a high-resolution, metric depth map via diffusion-based completion and DPT (Dense Prediction Transformer) refinement, providing robust Gaussian initialization and geometric constraints. Then, a DINO-guided view-consistent learning is introduced to augment Gaussian attributes with structural features, improving multi-view consistency. Finally, a diffusion-based Gaussian refinement module injects generative priors into an iterative optimization strategy, enhancing high-frequency geometric and appearance details within the Gaussian representation. Experiments on DTU, LLFF, and Mip-NeRF 360 show that D$^{3}$GS achieves consistent and substantial improvements over strong baselines, with ablation studies validating the effectiveness and complementary roles of each component.
This work proposes a planar prior guided 3D Gaussian Splatting framework for indoor scene reconstruction, and incorporates region-aware geometric optimization into the 3DGS pipeline, including planar region constraints, non-planar local geometric regularization, and global geometry supervision to jointly improve struct...
Liang-Ting Wang, Hong-Xing Qin· Proceedings of the 34th ACM...· 0 citations
Remote sensing novel view synthesis under sparse observations remains challenging due to insufficient geometric constraints and limited cross-view supervision. Existing Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) methods are prone to overfitting and face challenges of depth ambiguities, missing cross...
A lightweight gradient-driven point allocation strategy for depth-guided Gaussian splatting that bypasses traditional Structure-from-Motion initialization and incorporates a bidirectional re-projection check to eliminate transient noise and occlusions, ensuring geometric consistency is proposed.
Yu-Hao Shang, Zhen-Shen Qu, Guan-Yi Rong et al.· SAE technical paper series· 0 citations
Surface reconstruction under sparse-view settings remains challenging due to limited geometric cues. Volume rendering methods based on signed distance functions often produce over-smoothed surfaces, while 3D Gaussian Splatting (3DGS), though time-efficient, suffers from incomplete geometry due to the lack of reliable d...
Younghyun Noh, Minje Kim, Tae-Kyun Kim· 0 citations
Driving-scene reconstruction via feed-forward Gaussian splatting from sparse multi-view images remains challenging due to two factors. First, cross-camera photometric inconsistency, caused by variations in exposure, ISP, and illumination, can be incorrectly absorbed into Gaussian geometry. Second, unreliable local geom...
Yi-Fei Wang, Wei Tian, Wen-Hua Bao et al.· 1 citation· ⚡1
A 4D Gaussian Splatting framework for the Sparse-View Track of the SIGGRAPH Asia 2026 Volumetric Video Challenge, which requires dynamic scene reconstruction from only six cameras with wide baselines, ranking first overall in the Sparse-View Track.
Sheng-Qi Wang, Zheng-Xian Yang, Kai-Wen Tian et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.