Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

BRep Boundary and Junction Detection for CAD Reverse Engineering

Sk Aziz Ali, Mohammad Sadil Khan, Didier Stricker

TL;DR

This work tackles fast 3D reverse engineering by learning to map scans to Boundary-Representations (BRep) for subsequent CAD modeling. It introduces BRepDetNet, a dual-head network based on DGCNN that performs boundary and junction detection from 3D scans, augmented with a differentiable NMS loss to suppress false positives. The authors provide large-scale, per-vertex BRep annotations on the CC3D and ABC datasets and demonstrate superior boundary and junction detection performance, with strong cross-dataset generalization. The approach enables more reliable Scan-to-BRep pipelines, facilitating downstream BRep-to-CAD reconstruction and parameterized CAD design steps.

Abstract

In machining process, 3D reverse engineering of the mechanical system is an integral, highly important, and yet time consuming step to obtain parametric CAD models from 3D scans. Therefore, deep learning-based Scan-to-CAD modeling can offer designers enormous editability to quickly modify CAD model, being able to parse all its structural compositions and design steps. In this paper, we propose a supervised boundary representation (BRep) detection network BRepDetNet from 3D scans of CC3D and ABC dataset. We have carefully annotated the 50K and 45K scans of both the datasets with appropriate topological relations (e.g., next, mate, previous) between the geometrical primitives (i.e., boundaries, junctions, loops, faces) of their BRep data structures. The proposed solution decomposes the Scan-to-CAD problem in Scan-to-BRep ensuring the right step towards feature-based modeling, and therefore, leveraging other existing BRep-to-CAD modeling methods. Our proposed Scan-to-BRep neural network learns to detect BRep boundaries and junctions by minimizing focal-loss and non-maximal suppression (NMS) during training time. Experimental results show that our BRepDetNet with NMS-Loss achieves impressive results.

BRep Boundary and Junction Detection for CAD Reverse Engineering

TL;DR

This work tackles fast 3D reverse engineering by learning to map scans to Boundary-Representations (BRep) for subsequent CAD modeling. It introduces BRepDetNet, a dual-head network based on DGCNN that performs boundary and junction detection from 3D scans, augmented with a differentiable NMS loss to suppress false positives. The authors provide large-scale, per-vertex BRep annotations on the CC3D and ABC datasets and demonstrate superior boundary and junction detection performance, with strong cross-dataset generalization. The approach enables more reliable Scan-to-BRep pipelines, facilitating downstream BRep-to-CAD reconstruction and parameterized CAD design steps.

Abstract

In machining process, 3D reverse engineering of the mechanical system is an integral, highly important, and yet time consuming step to obtain parametric CAD models from 3D scans. Therefore, deep learning-based Scan-to-CAD modeling can offer designers enormous editability to quickly modify CAD model, being able to parse all its structural compositions and design steps. In this paper, we propose a supervised boundary representation (BRep) detection network BRepDetNet from 3D scans of CC3D and ABC dataset. We have carefully annotated the 50K and 45K scans of both the datasets with appropriate topological relations (e.g., next, mate, previous) between the geometrical primitives (i.e., boundaries, junctions, loops, faces) of their BRep data structures. The proposed solution decomposes the Scan-to-CAD problem in Scan-to-BRep ensuring the right step towards feature-based modeling, and therefore, leveraging other existing BRep-to-CAD modeling methods. Our proposed Scan-to-BRep neural network learns to detect BRep boundaries and junctions by minimizing focal-loss and non-maximal suppression (NMS) during training time. Experimental results show that our BRepDetNet with NMS-Loss achieves impressive results.
Paper Structure (13 sections, 7 equations, 5 figures, 2 tables)

This paper contains 13 sections, 7 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. BRep Data Annotation Features
  3. Annotation Quality Assessment
  4. B-Rep Boundary and Junction Detection
  5. Network Architecture of BRepDetNet
  6. BRep Boundary Detection
  7. B-Rep Junction Detection
  8. Experimental Results
  9. Baseline Models
  10. Comparison with The Baseline Models
  11. Ablation Study
  12. Limitation and Conclusion
  13. Acknowledgements

Figures (5)

  • Figure 1: The left figures relate a 3D scan with its corresponding BRep topology (color-coded by unique edge and different closed-loop ids). A list of BRep annotations per scan point (on the right) covers all attributes under CCLC.
  • Figure 2:
  • Figure 3: Our network architecture.
  • Figure 4: Visual results for boundary (left) and junction (right) prediction on ABC abc_dataset (top) and CC3D cc3d (bottom) dataset. The red points are the boundary points and the green points are the junction points. All the models have been trained on ABC abc_dataset dataset.
  • Figure 5: Visual results for boundary detection for BRepDetNet with and without NMS Loss. The red points are the boundary points. We observe that using NMS loss significantly reduces the number of false positives leading to higher precision in boundary detection.