MrNeRF (@janusch_patas)
2025-05-14 | โค๏ธ 236 | ๐ 32
[SIGGRAPH โ25] Monocular Online Reconstruction with Enhanced Detail Preservation
Abstract (excerpt): Our approach addresses two key challenges in monocular online reconstruction:
- Distributing Gaussians without relying on depth maps.
- Ensuring both local and global consistency in the reconstructed maps.
To achieve this, we introduce two key modules:
- Hierarchical Gaussian Management Module: For effective Gaussian distribution.
- Global Consistency Optimization Module: For maintaining alignment and coherence at all scales.
In addition, we present the Multi-level Occupancy Hash Voxels (MOHV), a structure that regularizes Gaussians to capture details across multiple levels of granularity. MOHV ensures accurate reconstruction of both fine and coarse geometries and textures, preserving intricate details while maintaining overall structural integrity.
Compared to state-of-the-art RGB-only and even RGB-D methods, our framework achieves superior reconstruction quality with high computational efficiency.
๋ฏธ๋์ด
๐ Related
Auto-generated - needs manual review