ESC: Escaping 2D Diffusion Ambiguity with Geometric Constraints for Consistent Novel View Synthesis
ESC is a geometry-aware diffusion framework for monocular novel view synthesis under drastic camera changes. It jointly predicts target RGB and depth, conditioned on the source image/depth and target camera rays via Plücker ray conditioning. A Projection Consistency Constraint (PCC) enforces bidirectional 3D depth re-projection consistency, and inference uses PCC-based best-of-N selection to return the most geometrically aligned sample.
