English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Surface reconstruction for point clouds is an important task in 3D computervision. Most of the latest methods resolve this problem by learning signeddistance functions from point clouds, which are limited to reconstructingclosed surfaces. Some other methods tried to represent open surfaces usingunsigned distance functions (UDF) which are learned from ground truthdistances. However, the learned UDF is hard to provide smooth distance fieldsdue to the discontinuous character of point clouds. In this paper, we proposeCAP-UDF, a novel method to learn consistency-aware UDF from raw point clouds.We achieve this by learning to move queries onto the surface with a fieldconsistency constraint, where we also enable to progressively estimate a moreaccurate surface. Specifically, we train a neural network to gradually inferthe relationship between queries and the approximated surface by searching forthe moving target of queries in a dynamic way. Meanwhile, we introduce apolygonization algorithm to extract surfaces using the gradients of the learnedUDF. We conduct comprehensive experiments in surface reconstruction for pointclouds, real scans or depth maps, and further explore our performance inunsupervised point normal estimation, which demonstrate non-trivialimprovements of CAP-UDF over the state-of-the-art methods.

CAP-UDF: Learning Unsigned Distance Functions Progressively from Raw Point Clouds with Consistency-Aware Field Optimization