3D Reconstruction from Images

VoxelMap provides two ways to go from 2D image → 3D model:

• ImageMap → voxelized volume

• ImageMesh → low-poly mesh (.obj)

—

Voxelized Image with ImageMap

from voxelmap import Model
import cv2, matplotlib.pyplot as plt

# load an image (grayscale topography)
model = Model(file="docs/img/land.png")

# blur for smoother levels
model.array = cv2.blur(model.array, (10, 10))

# map to voxel depth
model.array = model.ImageMap(depth=12)

# gradient coloring
from matplotlib import cm
model.colormap = cm.terrain
model.alphacm = 0.5
model.draw_mpl("linear", figsize=(8, 6))

➡ Produces a voxelized terrain.

—

Low-Poly Mesh with ImageMesh

model.ImageMesh(out_file="land.obj", L_sectors=15)
model.MeshView("land.obj", color="white", alpha=0.8)

➡ Produces a lightweight .obj mesh (interactive).