XScale-NVS: Cross-Scale Novel View Synthesis with Hash Featurized
Manifold
XScale-NVS: Cross-Scale Novel View Synthesis with Hash Featurized
Manifold
We propose XScale-NVS for high-fidelity cross-scale novel view synthesis of real-world large-scale scenes. Existing representations based on explicit surface suffer from discretization resolution or UV distortion, while implicit volumetric representations lack scalability for large scenes due to the dispersed weight distribution and surface ambiguity. In light of the above challenges, …