VoxCity

https://voxcity.readthedocs.io/en/latest/
Admin
VoxCity turns open building, land cover, canopy, and terrain data into a semantic 3D voxel model you...
Tokens:1,004,458
Snippets:4,678
Trust Score:9.6
Update:1 day ago
Show doc for...
Tokens:
Raw
### Setup and example area for demonstrations

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_data_sources.html

Imports VoxCity generator functions and defines a rectangle around Tokyo for use in examples.

```python
# Setup
from voxcity.generator import get_voxcity, auto_select_data_sources

# Example area for demonstrations
rectangle_vertices = [
    (139.760, 35.680),  # Tokyo area
    (139.760, 35.690),
    (139.770, 35.690),
    (139.770, 35.680)
]
```

--------------------------------

### Full import and export example

Source: https://voxcity.readthedocs.io/en/latest/guides/rhino_obj_import.html

Complete workflow: get a VoxCity model, import buildings from an OBJ file with anchor and move parameters, then export the result. Iterate on rotation/move and re-export to verify placement.

```python
from voxcity.generator import get_voxcity
from voxcity.importer import add_buildings_from_obj
from voxcity.exporter.obj import export_obj

vc = get_voxcity(rectangle_vertices, meshsize=2.0)
vc = add_buildings_from_obj(
    vc, "design.obj",
    anchor_lonlat=(139.7536, 35.6841),
    anchor_elevation=12.0,
    anchor_model_point=(0.0, 0.0, 0.0),
    rotation=0.0, move=(0.0, 0.0, 0.0), units="m",
)
export_obj(vc, "output", "with_imported_building")
```

--------------------------------

### Install VoxCity with pip

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_quickstart.html

Uncomment and run this cell to install VoxCity if it is not already installed. Skip if already installed.

```python
# Uncomment and run if you need to install VoxCity
# %pip install voxcity
```

--------------------------------

### Verify the installation

Source: https://voxcity.readthedocs.io/en/latest/installation.html

Imports voxcity and prints its version to confirm the installation.

```python
import voxcity
print(voxcity.__version__)
```

--------------------------------

### Install VoxCity and dependencies in fresh environment

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_solar.html

Installation commands for a fresh environment, with pip installs commented out. Includes VoxCity and additional dependencies: contextily, osmnx, geopandas, matplotlib, timezonefinder.

```python
# Installation (uncomment if running in a fresh environment)
# %pip install voxcity
# %pip install contextily osmnx geopandas matplotlib timezonefinder
```

--------------------------------

### Setup VoxCity and get building data

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_landmark.html

Import VoxCity modules, define a study area rectangle, and create a VoxCity object with specified data sources. Access the building GeoDataFrame from the object's extras.

```python
# %pip install voxcity

from voxcity.generator import get_voxcity
from voxcity.geoprocessor.draw import draw_rectangle_map_cityname
from voxcity.simulator.view import get_landmark_visibility_map

meshsize = 5
cityname = "Tokyo, Japan"

rectangle_vertices = [
    (139.760, 35.680),  # SW
    (139.760, 35.690),  # NW
    (139.770, 35.690),  # NE
    (139.770, 35.680)   # SE
]

city = get_voxcity(
    rectangle_vertices,
    meshsize=meshsize,
    building_source='OpenStreetMap',
    land_cover_source='OpenStreetMap',
    canopy_height_source='High Resolution 1m Global Canopy Height Maps',
    dem_source='DeltaDTM',
    output_dir='output/landmark_demo'
)

# Access building_gdf from the VoxCity object
building_gdf = city.extras.get('building_gdf', None)
len(building_gdf) if building_gdf is not None else 0
```

--------------------------------

### Install VoxCity and visualization dependencies

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_3d_visualization.html

Install the required packages for VoxCity and its visualization tools.

```bash
pip install voxcity plotly pyvista
```

--------------------------------

### Install VoxCity and system dependencies

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_basic.html

Installs the voxcity package and required system libraries (xvfb, libgl1-mesa-glx) for headless rendering. Run this in a Colab notebook.

```bash
#@title 1.1. Install voxcity
!pip install voxcity
!apt-get update
!apt-get install -y xvfb libgl1-mesa-glx
```

--------------------------------

### Install VoxCity with pip

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_landmark.html

Install the VoxCity package using pip. This is a prerequisite for running the tutorial.

```bash
pip install voxcity
```

--------------------------------

### Set up local development environment with virtualenvwrapper

Source: https://voxcity.readthedocs.io/en/latest/contributing.html

Creates a virtual environment, enters the project directory, and installs the package in development mode using setup.py develop.

```bash
$ mkvirtualenv voxcity
$ cd voxcity/
$ python setup.py develop
```

--------------------------------

### Example: setup_transformer

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/geoprocessor/utils/index.html

Shows how to set up a coordinate transformer between EPSG:4326 and EPSG:3857, then transform coordinates. The transformer uses always_xy=True for consistent (x,y) order.

```python
>>> transformer = setup_transformer("EPSG:4326", "EPSG:3857")
>>> x, y = transformer.transform(longitude, latitude)
```

--------------------------------

### Get raster bounding box with get_raster_bbox

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/geoprocessor/utils/index.html

Example of retrieving the bounding box of a raster file using get_raster_bbox. The example prints the bounds of the raster extent.

```python
>>> bbox = get_raster_bbox("elevation.tif")
>>> print(f"Raster extent: {bbox.bounds}")  # (minx, miny, maxx, maxy)
```

--------------------------------

### Get land cover classes for Urbanwatch source

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/utils/lc/index.html

Example usage of get_land_cover_classes to retrieve the RGB-to-class mapping for the 'Urbanwatch' source. The example shows that the tuple (255, 0, 0) maps to 'Building'.

```python
>>> classes = get_land_cover_classes("Urbanwatch")
>>> print(classes[(255, 0, 0)])  # Returns 'Building'
```

--------------------------------

### get_building_height_grid function definition and setup

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/generator/grids.html

This snippet shows the start of the get_building_height_grid function, including imports, rectangle normalization, Earth Engine initialization for the 'Open Building 2.5D Temporal' source, logging, and cache kwargs setup.

```python
[docs]
def get_building_height_grid(rectangle_vertices, meshsize, source, output_dir, building_gdf=None, **kwargs):
    from ..downloader.gee import (
        initialize_earth_engine,
        get_roi,
        save_geotiff_open_buildings_temporal,
        save_geotiff_dsm_minus_dtm,
    )

    if rectangle_vertices is not None:
        rectangle_vertices = normalize_rectangle_vertices(rectangle_vertices)

    ee_required_sources = {"Open Building 2.5D Temporal"}
    if source in ee_required_sources:
        initialize_earth_engine()

    quiet = kwargs.get('quiet', False)
    if not quiet:
        _logger.info("Creating Building Height grid\n ")
        _logger.info(f"Base data source: {source}")

    # Forward the on-disk download cache controls to any @cached_download
    # downloader calls below (they are consumed there, not by this function).
    _cache_kwargs = {
        "use_download_cache": kwargs.get("use_download_cache", True),
        "force_refresh": kwargs.get("force_refresh", False),
    }

    os.makedirs(output_dir, exist_ok=True)
```

--------------------------------

### prepare_cumulative_simulation_input example

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/simulator_gpu/solar/epw/index.html

Shows how to prepare solar simulation input data from an EPW file, returning azimuth, elevation, DNI, DHI, and location metadata, then printing location and daytime hours.

```python
>>> az, el, dni, dhi, loc = prepare_cumulative_simulation_input("weather.epw")
>>> print(f"Location: {loc.city}, {loc.country}")
>>> print(f"Daytime hours: {len(az)}")
```

--------------------------------

### convert_colormap_indices example

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/exporter/index.html

Doctest example showing how to convert a color map with arbitrary integer keys to sequential indices starting from 0. The output is logged via _logger.info and shows the new map with keys 0, 1, 2.

```python
>>> original = {5: [255, 0, 0], 10: [0, 255, 0], 15: [0, 0, 255]}
>>> new_map = convert_colormap_indices(original)
>>> _logger.info(new_map)
{0: [255, 0, 0], 1: [0, 255, 0], 2: [0, 0, 255]}
```

--------------------------------

### Static rendering with simulation overlays

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/visualizer/renderer.html

Demonstrates static rendering with a building simulation mesh (svf_mesh) and an output directory. Returns a list of (view_name, filepath) tuples.

```python
>>> visualize_voxcity(city, mode="static",
...                   building_sim_mesh=svf_mesh,
...                   output_directory="renders")
```

--------------------------------

### Setup: Get rectangle vertices for an area

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_data_sources.html

Initializes a rectangle around a center point (Marunouchi, Tokyo) using get_rectangle_vertices. This rectangle is used as input for data download functions.

```python
# Setup
from voxcity.geo.utils import get_rectangle_vertices

# Example area: Marunouchi, Tokyo
center = (139.7671, 35.6812)  # (lon, lat)
rectangle_vertices = get_rectangle_vertices(center, 200, 200)
```

--------------------------------

### Prepare cumulative simulation input and bin sun positions

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/simulator_gpu/solar/sky/index.html

Shows how to load weather data and bin sun positions to sky patches using prepare_cumulative_simulation_input and bin_sun_positions_to_patches. The example prints the number of active patches.

```python
>>> from palm_solar.epw import prepare_cumulative_simulation_input
>>> az, el, dni, dhi, loc = prepare_cumulative_simulation_input("weather.epw")
>>> binned = bin_sun_positions_to_patches(az, el, dni, dhi)
>>> print(f"Active patches: {(binned.hours_per_patch > 0).sum()}")
```

--------------------------------

### Interactive 3D visualization with Plotly

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_quickstart.html

Creates an interactive 3D viewer using Plotly. Requires plotly (pip install plotly). Use mode='interactive', show=False, and return_fig=True to get the figure, then call fig.show().

```python
# Interactive visualization using Plotly
# (Requires plotly: pip install plotly)
fig = visualize_voxcity(city, mode="interactive", show=False, return_fig=True)
fig.show()
```

--------------------------------

### UI setup for tree editor

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/geoprocessor/draw/edit_tree.html

Creates HTML style, input widgets for tree dimensions, buttons for adding/removing trees, and a status bar.

```python
    # --- UI ---
    style_html = HTML(generate_editor_css(
        "gm-tree-root",
        drawing_mode_class="tree-drawing-mode",
        delete_mode_class="tree-delete-mode",
    ))

    top_height_input = FloatText(
        value=10.0, description="Top (m):",
        layout=Layout(width="160px", height="24px"),
        style={"description_width": "62px"},
    )
    bottom_height_input = FloatText(
        value=4.0, description="Trunk (m):",
        layout=Layout(width="160px", height="24px"),
        style={"description_width": "62px"},
    )
    crown_diameter_input = FloatText(
        value=6.0, description="Dia. (m):",
        layout=Layout(width="160px", height="24px"),
        style={"description_width": "62px"},
    )
    fixed_prop_checkbox = Checkbox(
        value=True, description="Fixed proportion", indent=False,
        layout=Layout(width="auto", height="20px"),
    )

    add_click_button = Button(description="Click", button_style="success", layout=Layout(flex="1", height="26px"))
    add_area_button = ToggleButton(
        value=False, description="Area", button_style="",
        layout=Layout(flex="1", height="26px"), tooltip="Draw polygon to fill with trees",
    )
    remove_click_button = Button(description="Click", button_style="", layout=Layout(flex="1", height="26px"))
    remove_area_button = ToggleButton(
        value=False, description="Area", button_style="",
        layout=Layout(flex="1", height="26px"), tooltip="Draw polygon to remove trees & canopy inside",
    )

    status_bar = HTML(
        value="<div style='padding:3px 8px;border-radius:10px;font-size:10px;"
              "text-align:center;background:#f0f4ff;color:#1a73e8;'>Ready</div>"
    )
    hover_info = HTML("")
    _set_status = make_status_setter(status_bar, use_gm_class=False)

    add_row = HBox([add_click_button, add_area_button], layout=Layout(margin="0", gap="4px"))
    remove_row = HBox([remove_click_button, remove_area_button], layout=Layout(margin="0", gap="4px"))
    param_col = VBox(
        [top_height_input, bottom_height_input, crown_diameter_input, fixed_prop_checkbox],
        layout=Layout(margin="0", gap="0px"),
    )

    crown_diameter_input.layout.display = ""
```

--------------------------------

### Get patch info for Reinhart method

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/simulator_gpu/solar/sky/index.html

Shows how to retrieve information about a sky discretization method using get_patch_info. The example uses the 'reinhart' method with mf=4 and prints the method name and number of patches.

```python
>>> info = get_patch_info("reinhart", mf=4)
>>> print(f"{info['method']}: {info['n_patches']} patches")
```

--------------------------------

### GPURenderer initialization and custom lighting

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/visualizer/renderer_gpu.html

Initializes the GPURenderer with specified parameters and applies custom lighting if any lighting parameters are provided.

```python
renderer = GPURenderer(
    width=width,
    height=height,
    samples_per_pixel=samples_per_pixel,
    max_depth=max_depth,
    arch=arch
)

# Apply custom lighting if specified
if light_direction is not None or direct is not None or ambient is not None:
    current_direction = light_direction if light_direction is not None else light_direction_from_angles(220, 45)
    current_color = direct if direct is not None else (0.9, 0.9, 0.85)
    current_ambient = ambient if ambient is not None else (0.15, 0.15, 0.18)
    renderer.taichi_renderer.set_lighting(
        direction=current_direction,
        color=current_color,
        ambient=current_ambient,
        light_noise=0.3
    )
```

--------------------------------

### Initialize WGS84 Geod and perform inverse calculation

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/geoprocessor/utils/index.html

Example usage of initialize_geod to get a Geod object and then use its inv method to calculate forward azimuth, back azimuth, and distance between two points.

```python
>>> geod = initialize_geod()
>>> fwd_az, back_az, dist = geod.inv(lon1, lat1, lon2, lat2)
```

--------------------------------

### Region selection and output directory setup

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/utils/weather/onebuilding.html

Creates the output directory and builds a dictionary of regions to scan based on the 'region' parameter. Handles auto-detection, 'all', a single region, or a list of regions, raising ValueError for invalid selections.

```python
    try:
        Path(output_dir).mkdir(parents=True, exist_ok=True)

        regions_to_scan = {}
        def _add_selection(selection_name: str, mapping: Dict[str, str], out: Dict[str, str]):
            if selection_name in REGION_DATASET_GROUPS:
                for key in REGION_DATASET_GROUPS[selection_name]:
                    if key in KML_SOURCES:
                        out[key] = KML_SOURCES[key]
            elif selection_name in KML_SOURCES:
                out[selection_name] = KML_SOURCES[selection_name]
            else:
                valid = sorted(list(REGION_DATASET_GROUPS.keys()) + list(KML_SOURCES.keys()))
                raise ValueError(f"Invalid region/dataset: '{selection_name}'. Valid options include: {', '.join(valid)}")

        if region is None:
            detected_regions = detect_regions(longitude, latitude)
            if detected_regions:
                print(f"Auto-detected regions: {', '.join(detected_regions)}")
                for r in detected_regions:
                    _add_selection(r, KML_SOURCES, regions_to_scan)
            else:
                print("Could not determine region from coordinates. Scanning all regions.")
                regions_to_scan = dict(KML_SOURCES)
        elif isinstance(region, str):
            if region.lower() == "all":
                regions_to_scan = dict(KML_SOURCES)
            else:
                _add_selection(region, KML_SOURCES, regions_to_scan)
        else:
            for r in region:
                _add_selection(r, KML_SOURCES, regions_to_scan)
```

--------------------------------

### Quick start: Generate a voxel model with get_voxcity

Source: https://voxcity.readthedocs.io/en/latest/

Use the get_voxcity function to generate a voxel model for a specified area. The example uses a small area in Seattle with a mesh size of 5 meters. Sources are auto-selected based on location.

```python
from voxcity.generator import get_voxcity

# A small area in Seattle; sources auto-selected by location
voxcity = get_voxcity(rectangle_vertices, meshsize=5)
```

--------------------------------

### convert_colormap_indices function

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/exporter/obj.html

Converts a color map with arbitrary integer keys to sequential indices starting from 0, preserving RGB values. Logs the new map for debugging. Example usage shows input {5: [255,0,0], 10: [0,255,0], 15: [0,0,255]} producing {0: [255,0,0], 1: [0,255,0], 2: [0,0,255]}.

```python
[docs]
def convert_colormap_indices(original_map):
    """
    Convert a color map with arbitrary indices to sequential indices starting from 0.
    
    This function takes a color map with arbitrary integer keys and creates a new map
    with sequential indices starting from 0, maintaining the original color values.
    This is useful for ensuring consistent material indexing in OBJ files.
    
    Args:
        original_map (dict): Dictionary with integer keys and RGB color value lists.
            Each value should be a list of 3 integers (0-255) representing RGB colors.
        
    Returns:
        dict: New color map with sequential indices starting from 0.
            The values maintain their original RGB color assignments.
            
    Example:
        >>> original = {5: [255, 0, 0], 10: [0, 255, 0], 15: [0, 0, 255]}
        >>> new_map = convert_colormap_indices(original)
        >>> _logger.info(new_map)
        {0: [255, 0, 0], 1: [0, 255, 0], 2: [0, 0, 255]}
    """
    # Sort the original keys to maintain consistent ordering
    keys = sorted(original_map.keys())
    new_map = {}
    
    # Create new map with sequential indices
    for new_idx, old_idx in enumerate(keys):
        new_map[new_idx] = original_map[old_idx]
    
    # Print the new colormap for debugging/reference
    _logger.info("new_colormap = {")
    for key, value in new_map.items():
        original_key = keys[key]
        original_line = str(original_map[original_key])
        comment = ""
        if "#" in original_line:
            comment = "#" + original_line.split("#")[1].strip()
        _logger.info(f"    {key}: {value},  {comment}")
    _logger.info("}")
    
    return new_map
```

--------------------------------

### Import and use voxcity.simulator_gpu

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/simulator_gpu/index.html

Shows how to import the solar and visibility submodules and call their main functions. The example uses get_global_solar_irradiance_using_epw for solar radiation and get_view_index with mode='green' for view analysis.

```python
from voxcity.simulator_gpu import solar, visibility

# Solar radiation
irradiance = solar.get_global_solar_irradiance_using_epw(voxcity, …)

# View analysis
gvi = visibility.get_view_index(voxcity, mode='green')
```

--------------------------------

### Helper function _require_trimesh

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/exporter/obj.html

Raises ConfigurationError if trimesh is not installed, instructing to install via 'pip install trimesh'.

```python
def _require_trimesh():
    if not _HAS_TRIMESH:
        raise ConfigurationError(
            "This operation requires the optional dependency 'trimesh'. "
            "Install it via 'pip install trimesh'."
        )
```

--------------------------------

### Building solar irradiance overlay example

Source: https://voxcity.readthedocs.io/en/latest/_modules/voxcity/visualizer/renderer_gpu.html

Renders with a building simulation mesh for solar irradiance, using the 'global' value name and 'magma' colormap.

```python
>>> building_mesh = get_building_solar_irradiance(city, ...)
>>> img = visualize_voxcity_gpu(city, building_sim_mesh=building_mesh, 
...                             building_value_name='global', building_colormap='magma')
```

--------------------------------

### Canopy Height Parameters Example

Source: https://voxcity.readthedocs.io/en/latest/examples/demo_data_sources.html

Defines canopy height configuration parameters including minimum canopy height, trunk height ratio, and static tree height. Prints the configuration options.

```python
# Canopy height parameters
canopy_params = {
    "min_canopy_height": 2.0,      # Ignore trees shorter than this (meters)
    "trunk_height_ratio": 0.59,    # Trunk height / total height
    "static_tree_height": 10.0     # Default height for 'Static' source
}

print("Canopy configuration options:")
for param, value in canopy_params.items():
    print(f"  {param}: {value}")
```

--------------------------------

### Install with conda + pip (recommended)

Source: https://voxcity.readthedocs.io/en/latest/installation.html

Creates a conda environment named 'voxcity' with Python 3.12, installs GDAL and timezonefinder from conda-forge, then installs voxcity via pip.

```bash
conda create --name voxcity python=3.12
conda activate voxcity
conda install -c conda-forge gdal timezonefinder
pip install voxcity
```

--------------------------------

### Static rendering with simulation overlays: visualize_voxcity with output_directory

Source: https://voxcity.readthedocs.io/en/latest/autoapi/voxcity/visualizer/renderer/index.html

Usage example for static rendering mode with building simulation mesh and output directory.

```python
>>> visualize_voxcity(city, mode="static", …
... building_sim_mesh=svf_mesh, …
... output_directory="renders")
```
Tokens:
Raw