MonoInstance: Enhancing Monocular Priors via Multi-view Instance Alignment for Neural Rendering and Reconstruction
This addresses the challenge of effectively leveraging monocular cues in multi-view neural rendering for 3D reconstruction and novel view synthesis, representing an incremental advancement over existing methods.
The paper tackles the problem of inconsistent monocular depth predictions across multiple views in neural rendering tasks, proposing MonoInstance which aligns instance depths in 3D space and introduces constraints for unreliable areas, resulting in significant performance improvements in reconstruction and novel view synthesis across benchmarks.
Monocular depth priors have been widely adopted by neural rendering in multi-view based tasks such as 3D reconstruction and novel view synthesis. However, due to the inconsistent prediction on each view, how to more effectively leverage monocular cues in a multi-view context remains a challenge. Current methods treat the entire estimated depth map indiscriminately, and use it as ground truth supervision, while ignoring the inherent inaccuracy and cross-view inconsistency in monocular priors. To resolve these issues, we propose MonoInstance, a general approach that explores the uncertainty of monocular depths to provide enhanced geometric priors for neural rendering and reconstruction. Our key insight lies in aligning each segmented instance depths from multiple views within a common 3D space, thereby casting the uncertainty estimation of monocular depths into a density measure within noisy point clouds. For high-uncertainty areas where depth priors are unreliable, we further introduce a constraint term that encourages the projected instances to align with corresponding instance masks on nearby views. MonoInstance is a versatile strategy which can be seamlessly integrated into various multi-view neural rendering frameworks. Our experimental results demonstrate that MonoInstance significantly improves the performance in both reconstruction and novel view synthesis under various benchmarks.