Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 5, 2026ACM Transactions on GraphicsOpen Access

Points as Tori: Fast Pointwise Signed Distance for Point Clouds

View Full Paper
Ask AI
Bookmark
Share

Authors

NFNicole C. FengIGIoannis GkioulekasKCKeenan Crane

Discussion

Loading...

Member takes

Overview

Randomized trial shows improved signed distance computation in point clouds, indicating efficient surface reconstruction methods.

Key Points

  • This research aims to provide a fast method for computing signed distance to point clouds by reconstructing shapes using tori.
  • Developed an analytical parameterization for querying signed distance
  • Fitted tori using a pre-trained neural network for curvature and shift parameters
  • Applied the method to various inputs: photogrammetry, meshes, 3D Gaussians, neural implicits
  • Demonstrated reduced computation time without global optimization
  • Enabled direct application of point clouds in various operations like offsets and visualizations
  • Unified signed distance theory with classical reconstruction methods, enhancing efficiency

Cite This Study

Feng et al. (2026) studied this question.

synapsesocial.com/papers/6a49f68df5d1d45b28800d6chttps://doi.org/10.1145/3811385
View Full Paper
Ask AI
Bookmark
Share