SurfMesh - A Surface Meshing Python Library

SurfMesh is a Python library for generating structured 3D surface meshes of primitive shapes, with a strong focus on quadrilateral-dominant (quad) meshing. The meshes are particularly suited for visualization and Boundary Element Method (BEM) simulations.

⚠️ This project is currently under active development.

🎯 Objective

This library aims to provide a minimal, intuitive interface for constructing quad-based surface meshes of primitive solids.

Use cases include:

• Geometry visualization
• Boundary Element Methods (BEM)
• Educational tooling
• Preprocessing for surface-based solvers

⚙️ Requirements

• Python: >= 3.10
• Dependencies:
  • numpy>=1.24
  • Optional (for examples and visualization):
    • ipykernel
    • jupyterlab
    • matplotlib

🚀 Installation

To install stable version via PyPI:

pip install surfmesh

For the latest development version via Git:

pip install git+https://github.com/ckesanapalli/surface-mesher.git

🧱 Basic Usage

Below are examples of how to use the library to generate and visualize meshes.

from pathlib import Path

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.patches import Polygon
from matplotlib.collections import PatchCollection
from mpl_toolkits.mplot3d.art3d import Poly3DCollection

import surfmesh as sm

plot_res = (4, 4)

1. Mesh Between Two Edges

# Define two edges
x = np.linspace(0, np.pi / 2, 20)
edge1 = np.array([x[2:], np.sin(x[2:])])
edge2 = np.array([x[:-2], np.exp(x[:-2])])

# Generate the mesh
radial_resolution = 10
mesh = sm.mesh_between_edges([edge1, edge2], radial_resolution)

print(f"Generated mesh with {mesh.shape[0]} quadrilateral faces.")

fig, ax = plt.subplots(figsize=plot_res)
collection = PatchCollection(map(Polygon, mesh), facecolor='lightblue', edgecolor='k', linewidth=0.3)
ax.add_collection(collection)
ax.set_xlim(mesh[:, :, 0].min() - 0.1, mesh[:, :, 0].max() + 0.1)
ax.set_ylim(mesh[:, :, 1].min() - 0.1, mesh[:, :, 1].max() + 0.1)
ax.set_title("Mesh Between Edges")
plt.show()

Generated mesh with 170 quadrilateral faces.

2. Radial Disk Mesh

# Parameters for the radial disk mesh
radius = 1.0
radial_resolution = 10
segment_resolution = 20

# Generate the radial disk mesh
radial_mesh = sm.disk_mesher_radial(radius, radial_resolution, segment_resolution)

print(f"Generated radial disk mesh with {radial_mesh.shape[0]} quadrilateral faces.")

fig, ax = plt.subplots(figsize=plot_res)
patches = [Polygon(face, closed=True) for face in radial_mesh]
collection = PatchCollection(patches, facecolors="lightblue", edgecolors="k", alpha=0.7)
ax.add_collection(collection)

ax.set_xlim(-radius, radius)
ax.set_ylim(-radius, radius)
ax.set_aspect("equal")
ax.set_title("Radial Disk Mesh")
plt.show()

Generated radial disk mesh with 200 quadrilateral faces.

3. Square-Centered Disk Mesh

radius = 1.0
square_resolution = 5
radial_resolution = 10
square_side_radius_ratio = 0.5

# Generate the square-centered disk mesh
square_centered_mesh = sm.disk_mesher_square_centered(radius, square_resolution, radial_resolution, square_side_radius_ratio)

print(f"Generated square-centered disk mesh with {square_centered_mesh.shape[0]} quadrilateral faces.")

fig, ax = plt.subplots(figsize=plot_res)
collection = PatchCollection(map(Polygon, square_centered_mesh), facecolors="lightgreen", edgecolors="k", alpha=0.7)
ax.add_collection(collection)

ax.set_xlim(-radius, radius)
ax.set_ylim(-radius, radius)
ax.set_aspect("equal")
ax.set_title("Square-Centered Disk Mesh")
plt.show()

Generated square-centered disk mesh with 225 quadrilateral faces.

4. Cuboid Mesh using Explicit Coordinates

# Define coordinate arrays for a cuboid
x_coords = [0.0, 1.0, 2.0]
y_coords = [0.0, 1.0, 2.0]
z_coords = [0.0, 0.5, 1.0]

# Generate the cuboid surface mesh
faces = sm.cuboid_mesher(x_coords, y_coords, z_coords)

print(f"Generated {faces.shape[0]} quadrilateral faces.")
print(faces.shape)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")

# Add each quad to the 3D plot
poly = Poly3DCollection(faces, facecolors="skyblue", edgecolors="k", alpha=0.7)
ax.add_collection3d(poly)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
ax.set_title("Cuboid Surface Mesh")
plt.tight_layout()
plt.show()

Generated 24 quadrilateral faces.
(24, 4, 3)

5. Cuboid Mesh using Resolution

# Generate a cuboid mesh with resolution
length, width, height = 2.0, 1.0, 1.0
resolution = (4, 2, 2)

mesh = sm.cuboid_mesher_with_resolution(length, width, height, resolution=resolution)

print(f"Generated cuboid with {mesh.shape[0]} quadrilateral faces.")
print(mesh.shape)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")

poly = Poly3DCollection(mesh, facecolors="lightgreen", edgecolors="k", alpha=0.6)
ax.add_collection3d(poly)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
ax.set_title("Cuboid Mesh with Resolution")
plt.tight_layout()
plt.show()

Generated cuboid with 40 quadrilateral faces.
(40, 4, 3)

6. Revolve a Curve Along a Circular Path

# Sample 2D curve coordinates
x = np.linspace(0, 1, 20)
z = x ** 2  # Example curve (parabola)

curves = np.array([x, z]).T
# Revolve the curve
segment_resolution = 20
faces = sm.circular_revolve(curves, segment_resolution, start_angle=0, end_angle=2*np.pi)

# Plotting
fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection='3d')
ax.add_collection3d(Poly3DCollection(faces, facecolors='g', linewidths=1, alpha=0.5))
ax.set_xlim(-x.max(), x.max())
ax.set_ylim(-x.max(), x.max())
ax.set_zlim(z.min(), z.max())
ax.set_xlabel('X-axis')
ax.set_ylabel('Y-axis')
ax.set_zlabel('Z-axis')
plt.show()

7. Revolve a Curve Along a Custom Path

x = np.linspace(1, 10, 100)
z = np.log(x)
main_curve = np.array([x, z]).T

angle_rad = np.linspace(0, 4*np.pi, 100)
radius = angle_rad/10
revolve_path = np.array([angle_rad, radius]).T

revolved_mesh = sm.revolve_curve_along_path(main_curve, revolve_path)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection='3d')
ax.add_collection3d(Poly3DCollection(revolved_mesh, alpha=0.5))
ax.set_xlim(-x.max(), x.max())
ax.set_ylim(-x.max(), x.max())
ax.set_zlim(z.min(), z.max())
plt.show()

8. Generate a Radial Cylinder Mesh

radius = 1.0
height = 2.0
radial_resolution = 8
segment_resolution = 16
height_resolution = 10

# Generate the cylinder mesh
mesh = sm.cylinder_mesher_radial(radius, height, radial_resolution, segment_resolution, height_resolution)

print(f"Generated a Radial Cylinder Mesh {mesh.shape[0]}.")
print(mesh.shape)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")

poly = Poly3DCollection(mesh, facecolors="lightgreen", edgecolors="k", alpha=0.6)
ax.add_collection3d(poly)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
ax.set_title("Radial Cylinder Mesh")
plt.tight_layout()
plt.show()

Generated a Radial Cylinder Mesh 416.
(416, 4, 3)

9. Generate a Square-Centered Cylinder Mesh

radius = 1.0
height = 2.0
radial_resolution = 8
half_square_side_resolution = 4
height_resolution = 10

# Generate the cylinder mesh
mesh = sm.cylinder_mesher_square_centered(radius, height, radial_resolution, half_square_side_resolution, height_resolution)

print(f"Generated a Square-Centered Cylinder Mesh.")
print(mesh.shape)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")

poly = Poly3DCollection(mesh, facecolors="lightgreen", edgecolors="k", alpha=0.6)
ax.add_collection3d(poly)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
ax.set_title("Square-Centered Cylinder Mesh")
plt.tight_layout()
plt.show()

Generated a Square-Centered Cylinder Mesh.
(960, 4, 3)

10. Generate a Sphere Mesh Using Cube Projection

mesh = sm.sphere_mesher_from_projection(radius=1.0, resolution=10)

print(f"Generated a Sphere Mesh from Cube Projection.")
print(mesh.shape)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")

poly = Poly3DCollection(mesh, facecolors="lightgreen", edgecolors="k", alpha=0.6)
ax.add_collection3d(poly)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
ax.set_title("Sphere Mesh from Cube Projection")
plt.tight_layout()
plt.show()

Generated a Sphere Mesh from Cube Projection.
(600, 4, 3)

11. Generate a Sphere Mesh Using Radial Divisions

radius = 1.0
radial_resolution = 20
segment_resolution = 20
mesh = sm.sphere_mesher_from_radial(radius, radial_resolution, segment_resolution)

print(f"Generated a Radial Sphere Mesh.")
print(mesh.shape)

fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")

poly = Poly3DCollection(mesh, facecolors="lightgreen", edgecolors="k", alpha=0.6)
ax.add_collection3d(poly)

ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
ax.set_title("Sphere Mesh from Radial Sphere")
plt.tight_layout()
plt.show()

Generated a Radial Sphere Mesh.
(400, 4, 3)

12. Extract Faces and Vertices from the mesh

radius = 1.0
radial_resolution = 20
segment_resolution = 20
mesh = sm.sphere_mesher_from_radial(radius, radial_resolution, segment_resolution)

vertices, faces = sm.extract_vertices_faces(mesh.round(6))

print(f"Generated a Radial Sphere Mesh with {faces.shape[0]} faces and {vertices.shape[0]} vertices.")
print(f"Vertices shape: {vertices.shape}, Faces shape: {faces.shape}")
print(f"First 5 vertices:\n{vertices[:5]}")
print(f"First 5 faces:\n{faces[:5]}")

Generated a Radial Sphere Mesh with 400 faces and 382 vertices.
Vertices shape: (382, 3), Faces shape: (400, 4)
First 5 vertices:
[[-1.        0.        0.      ]
 [-0.987688  0.       -0.156434]
 [-0.987688  0.        0.156434]
 [-0.951057 -0.309017  0.      ]
 [-0.951057  0.       -0.309017]]
First 5 faces:
[[233 235 190 190]
 [269 278 235 233]
 [295 297 278 269]
 [315 328 297 295]
 [337 339 328 315]]

13. Create a Curvilinear Mesh from Four Edges

Generate a structured quadrilateral mesh over a 4-sided curvilinear panel using a bilinearly-blended Coons patch (transfinite interpolation).

Given four boundary curves -- bottom, top, left, right -- each parameterized by t in [0, 1], this module fills in the interior surface points and the quad face connectivity needed to render the panel (e.g. with mpl_toolkits.mplot3d.art3d.Poly3DCollection)

def bottom(t: np.ndarray) -> np.ndarray:
    return np.c_[t, np.zeros_like(t), 0.3 * np.sin(np.pi * t)]

def top(t: np.ndarray) -> np.ndarray:
    return np.c_[t, 2 * np.ones_like(t), 0.5 * np.sin(np.pi * t) + 0.4]

def left(t: np.ndarray) -> np.ndarray:
    return np.c_[np.zeros_like(t), 2 * t, 0.4 * t]

def right(t: np.ndarray) -> np.ndarray:
    return np.c_[np.ones_like(t), 2 * t**5, 0.3 * np.sin(np.pi * t) + 0.4 * t]

grid, faces = sm.coons_patch(bottom, top, left, right, nu=24, nv=16)
fig = plt.figure(figsize=plot_res)
ax = fig.add_subplot(111, projection="3d")
ax.add_collection3d(Poly3DCollection(faces, facecolors="lightgreen", edgecolors="k", alpha=0.6))
pts = grid.reshape(-1, 3)
ax.set_xlim(pts[:, 0].min(), pts[:, 0].max())
ax.set_ylim(pts[:, 1].min(), pts[:, 1].max())
ax.set_zlim(pts[:, 2].min(), pts[:, 2].max())
ax.set_xlabel("X")
ax.set_ylabel("Y")
ax.set_zlabel("Z")
plt.tight_layout()
plt.show()

Citation

If you use this library in your research, please consider citing the following citation: CITATION.bib

from urllib.request import urlopen
from pathlib import Path

url = "https://zenodo.org/records/15298588/export/bibtex"
content = urlopen(url).read().decode("utf-8")
Path("../CITATION.bib").write_text(content, encoding="utf-8")
print(content)

@software{chaitanya_kesanapalli_2025_15298588,
  author       = {Chaitanya Kesanapalli},
  title        = {SurfMesh},
  month        = apr,
  year         = 2025,
  publisher    = {Zenodo},
  version      = {v0.2},
  doi          = {10.5281/zenodo.15298588},
  url          = {https://doi.org/10.5281/zenodo.15298588},
  swhid        = {swh:1:dir:17d13ade48cf763577a55d76cbd69c3ebfda8fbb
                   ;origin=https://doi.org/10.5281/zenodo.15298035;vi
                   sit=swh:1:snp:81a5188f930fc00f53a38853ff78cf140217
                   aafc;anchor=swh:1:rel:f2b2d3966583101539faa625db75
                   16e5e1d6393d;path=ckesanapalli-surfmesh-005395d
                  },
}

📌 Roadmap

• Cuboid surface mesh generation
• Disk face mesh generation
• Revolve curve mesh generation
• Cylinder, and sphere support
• Curvilinear mesh
• STL/PLY export support
• Mesh visualization utilities
• Export to BEM-compatible formats