
The core message of this paper is to propose a geometrically consistent cost aggregation scheme that leverages local geometric smoothness and surface normals to better utilize adjacent geometries, leading to improved multi-view stereo reconstruction performance.


coremsg

geometrically-consistent-cost-aggregation-for-efficient-multi-view-stereo-reconstruction


Geometrically Consistent Cost Aggregation for Efficient Multi-View Stereo Reconstruction


title_rewrite


CHOSEN, a simple yet flexible, robust and effective multi-view depth refinement framework, iteratively re-samples and selects the best depth hypotheses using contrastive learning, and automatically adapts to different metric or intrinsic scales determined by the capture system.


contrastive-hypothesis-selection-for-robust-and-accurate-multi-view-depth-refinement


Contrastive Hypothesis Selection for Robust and Accurate Multi-View Depth Refinement
