Search papers, labs, and topics across Lattice.
The paper introduces G$^2$ARD-GS, a geometry-guided distillation method that efficiently converts dense LiDAR maps into compact representations suitable for 3D Gaussian Splatting (3DGS). By progressively consolidating a Gaussian prior into surface-aware representatives and employing anchor constraints during recovery, the method maintains essential local surface details while significantly reducing the number of primitives. Experimental results on MatrixCity demonstrate that G$^2$ARD-GS achieves superior performance in image quality metrics (PSNR, SSIM, LPIPS) across various compression budgets, outperforming existing methods like PUP by notable margins.
G$^2$ARD-GS achieves up to 30x compression while enhancing image quality and preserving geometric fidelity, setting a new standard for 3D scene representation.
Dense colored LiDAR maps provide accurate city-scale geometry, but lifting them into 3D Gaussian Splatting (3DGS) retains millions of primitives, making the resulting models costly to store, transmit, render, and adapt. Aggressive primitive reduction alleviates this burden, but can remove the local surface support needed for stable novel-view synthesis and downstream geometric use. We introduce G$^2$ARD-GS, a geometry-guided distillation method that converts a dense Gaussian prior instantiated either as a training-free point-cloud lift or a trained GS model into a compact, reusable representation. G$^2$ARD-GS progressively consolidates the prior into surface-aware representatives, then recovers appearance on the resulting fixed topology under construction-time anchor constraints, with no primitives added or removed during recovery. Under limited supervision, geometry-aware view selection allocates the available view budget. On MatrixCity, G$^2$ARD-GS achieves the best PSNR, SSIM, and LPIPS across matched $5\times$--$30\times$ compression budgets, outperforming PUP by $3.2$--$6.8$,dB in PSNR. When reused as frozen geometry, the compact model improves off-trajectory appearance adaptation by $3.7$--$4.9$,dB over PUP 3D-GS and preserves image-to-model registration accuracy on Cambridge KingsCollege at $30\times$ compression. Project page: https://patrick1159.github.io/gardGS-page/.
