Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025IEEE Transactions on Pattern Analysis and Machine IntelligenceOpen Access

Probabilistic Directed Distance Fields for Ray-Based Shape Representations

View Full Paper
Ask AI
Bookmark
Share

Authors

TATristan Aumentado‐ArmstrongUniversity of TorontoSTStavros TsogkasSupélecSDSven DickinsonUniversity of Toronto

Discussion

Loading...

Member takes

Implication

This work demonstrates directed distance fields reduce issues in implicit representations, suggesting improved rendering for 3D models.

Key Points

  • Probabilistic directed distance fields enable more efficient differentiable rendering in 3D shape representations.
  • The approach allows obtaining depth with a single forward pass per pixel, enhancing geometric fidelity.
  • Directed distance fields showcase strong performance in single-shape fitting and 3D reconstruction tasks.
  • Theoretical investigation confirms that a small set of field properties ensures view consistency in the representation.

Cite This Study

Aumentado‐Armstrong et al. (2025) studied this question.

synapsesocial.com/papers/68c1a40f54b1d3bfb60dec20https://doi.org/10.1109/tpami.2025.3594225
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1SDFDiff: Differentiable Rendering of Signed Distance Fields for 3D Shape Optimization2020 · 212 citations
  2. 2Advances in Neural Rendering2022 · 337 citations
  3. 3KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs2021 · 676 citations
  4. 4PlenOctrees for Real-time Rendering of Neural Radiance Fields2021 · 856 citations
  5. 5Self-Supervised 2D Image to 3D Shape Translation with Disentangled Representations2020 · 7 citations