用混合隐变量分离几何与外观，用更少点实现更高精度的3D重建。
Hybrid Latents: Geometry-Appearance-Aware Surfel Splatting
- 在高斯点中引入频域隐变量，解耦几何与外观表示
- 仅用1/10数量的高斯点就达到领先重建精度
- 适合需要高效3D重建的应用，如AR/VR和机器人导航
我们提出一种混合高斯-哈希网格辐射场表示方法，用于从多视角图像重建2D高斯场景模型。与NeST splatting类似，该方法降低了基于NeRF模型中常见的几何与外观纠缠问题，但额外在每个高斯点上加入隐变量特征，并结合哈希网格特征，引导优化器将低频与高频场景成分分离。这种显式的频域分解减少了高频纹理对几何误差的补偿倾向。通过强制高斯点具有陡峭的不透明度衰减，进一步强化了几何与外观的分离，提升了几何重建精度和渲染效率。最后，结合概率剪枝与稀疏性诱导的BCE不透明度损失，可关闭冗余高斯点，得到最小必要高斯集合以表征场景。在合成与真实世界数据集上的实验表明，相比现有高斯基新视角生成方法，本方法以一个数量级更少的基元实现了更高的重建保真度。
原文摘要 · Abstract (English)
We introduce a hybrid Gaussian-hash-grid radiance representation for reconstructing 2D Gaussian scene models from multi-view images. Similar to NeST splatting, our approach reduces the entanglement between geometry and appearance common in NeRF-based models, but adds per-Gaussian latent features alongside hash-grid features to bias the optimizer toward a separation of low- and high-frequency scene components. This explicit frequency-based decomposition reduces the tendency of high-frequency texture to compensate for geometric errors. Encouraging Gaussians with hard opacity falloffs further strengthens the separation between geometry and appearance, improving both geometry reconstruction and rendering efficiency. Finally, probabilistic pruning combined with a sparsity-inducing BCE opacity loss allows redundant Gaussians to be turned off, yielding a minimal set of Gaussians sufficient to represent the scene. Using both synthetic and real-world datasets, we compare against the state of the art in Gaussian-based novel-view synthesis and demonstrate superior reconstruction fidelity with an order of magnitude fewer primitives.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。