AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Continuous Implicit Representation for Near-Periodic Patterns

Bowei Chen, Tiancheng Zhi, Martial Hebert, Srinivasa G. Narasimhan

ECCV 2022 · 3D vision › Face and shape · approach: Implicit neural representation design · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add periodicity-aware input warping and periodicity prior (ablation)NPP-Net · NRTDBLPIPS: 0.429 -> 0.188Improved
add periodicity-aware input warping and periodicity prior (ablation)NPP-Net · DTDLPIPS: 0.468 -> 0.249Improved
add periodicity-guided patch loss to pixel loss (pixel only to full) (ablation)NPP-Net · NRTDBLPIPS: 0.308 -> 0.188Improved
add pixel loss to patch loss (patch only to full) (ablation)NPP-Net · NRTDBLPIPS: 0.322 -> 0.188Improved
sample patches by periodicity instead of randomly (ablation)NPP-Net · NRTDBLPIPS: 0.216 -> 0.188Improved
sample patches by periodicity instead of randomly (ablation)NPP-Net · FacadeLPIPS: 0.316 -> 0.263Improved
add periodicity proposal with Top-3 candidates and offsets to initial pipeline (ablation)NPP-Net · NRTDBLPIPS: 0.213 -> 0.188Improved
use Top-3 instead of Top-1 periodicities (with offsets) (ablation)NPP-Net · NRTDBLPIPS: 0.205 -> 0.188Improved
use Top-3 instead of Top-5 periodicities (with offsets) (ablation)NPP-Net · NRTDBLPIPS: 0.197 -> 0.188Improved
add fine-level offsets to Top-3 periodicities (ablation)NPP-Net · NRTDBLPIPS: 0.210 -> 0.188Improved
NPP-Net versus Lama on non-strictly periodic facadesNPP-Net · FacadeLPIPS: 0.207 vs 0.263Hurt
NPP-Net versus Lama trained on large datasetsNPP-Net · NRTDBLPIPS: 0.196 -> 0.188Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy