AI Paper Atlas

Research map / 3D vision / Face and shape

Neural Field-Based 3D Surface Reconstruction of Microstructures from Multi-Detector Signals in Scanning Electron Microscopy

Shuo Chen, Yijin Li, Xi Zheng, Guofeng Zhang

CVPR 2026 · 3D vision › Face and shape · approach: Neural implicit surface reconstruction · 1 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 learnable BSE forward model instead of direct gradient supervision (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 135.61 -> 17.48Improved
replace analytical emission model with fourth-order polynomial reflectance term (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)BSE model error: 7.16 -> 0.27Improved
replace analytical emission model with fourth-order polynomial reflectance term (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 19.96 -> 17.48Improved
use independent learnable parameters per detector quadrant (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 19.90 -> 17.48Improved
use independent learnable parameters per detector quadrant (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)BSE model error: 1.35 -> 0.27Improved
add iterative shadow separation (third training stage) (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 29.38 -> 17.48Improved
add iterative shadow separation (third training stage) (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)BSE model error: 0.61 -> 0.27Improved
fuse BSE cues in a neural field instead of single-view photometric stereo integration (ablation)NFH-SEM · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 512.22 -> 17.48Improved
supervise with direct gradients instead of refining coarse model (ablation)NFH-SEM w/o BSE-F · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 25.11 -> 135.61Hurt
remove shadow masking (ablation)NFH-SEM w/o S-Mask · Simulated TPL dataset (Wukong, Lucy, Lion avg)Chamfer distance (nm): 25.11 -> 29.38Hurt

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