Radiant Triangle Soup with Soft Connectivity Forces for 3D Reconstruction and Novel View Synthesis
This work addresses the problem of achieving accurate geometry in 3D scene representations for computer vision and graphics researchers, though it appears incremental as it builds on existing triangle-based methods with a novel regularization technique.
The paper tackles 3D reconstruction and novel view synthesis by introducing an inference-time scene optimization algorithm using triangle soup with soft connectivity forces, resulting in improved geometric accuracy compared to state-of-the-art methods while maintaining visual fidelity.
We introduce an inference-time scene optimization algorithm utilizing triangle soup, a collection of disconnected translucent triangle primitives, as the representation for the geometry and appearance of a scene. Unlike full-rank Gaussian kernels, triangles are a natural, locally-flat proxy for surfaces that can be connected to achieve highly complex geometry. When coupled with per-vertex Spherical Harmonics (SH), triangles provide a rich visual representation without incurring an expensive increase in primitives. We leverage our new representation to incorporate optimization objectives and enforce spatial regularization directly on the underlying primitives. The main differentiator of our approach is the definition and enforcement of soft connectivity forces between triangles during optimization, encouraging explicit, but soft, surface continuity in 3D. Experiments on representative 3D reconstruction and novel view synthesis datasets show improvements in geometric accuracy compared to current state-of-the-art algorithms without sacrificing visual fidelity.