### Build PyCOLMAP from source on Linux/macOS

Source: https://colmap.github.io/pycolmap/index.html

Build PyCOLMAP from source on Linux and macOS after installing COLMAP from source. Ensure COLMAP is installed first.

```bash
python -m pip install .
```

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

### Install PyCOLMAP using pip

Source: https://colmap.github.io/pycolmap/index.html

Install pre-built wheels for PyCOLMAP on Linux, macOS, and Windows using pip. For GPU acceleration on Linux, consider the pycolmap-cuda12 package.

```bash
pip install pycolmap
```

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

### FeatureMatcher.match_guided

Source: https://colmap.github.io/pycolmap/pycolmap.html

Performs guided matching using existing two-view geometry.

```APIDOC
## match_guided

### Description
Perform guided matching using existing two-view geometry. Updates the two_view_geometry in-place.

### Parameters
#### Request Body
- **max_error** (SupportsFloat) - Required - Maximum allowed error.
- **keypoints1** (pycolmap.FeatureKeypoints) - Required - Keypoints from the first image.
- **descriptors1** (pycolmap.FeatureDescriptors) - Required - Descriptors from the first image.
- **camera1** (pycolmap.Camera) - Required - Camera model for the first image.
- **keypoints2** (pycolmap.FeatureKeypoints) - Required - Keypoints from the second image.
- **descriptors2** (pycolmap.FeatureDescriptors) - Required - Descriptors from the second image.
- **camera2** (pycolmap.Camera) - Required - Camera model for the second image.
- **two_view_geometry** (pycolmap.TwoViewGeometry) - Required - The two-view geometry to update.
```

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

### LightGlueONNXMatchingOptions Initialization

Source: https://colmap.github.io/pycolmap/pycolmap.html

Options for configuring LightGlue matching using ONNX models. Supports initialization with default values, a dictionary, or keyword arguments.

```python
class pycolmap.LightGlueONNXMatchingOptions(_* args_, _** kwargs_)
    
Overloaded function.
  1. `__init__(self: pycolmap.LightGlueONNXMatchingOptions) -> None`
  2. `__init__(self: pycolmap.LightGlueONNXMatchingOptions, arg0: dict) -> None`
  3. `__init__(self: pycolmap.LightGlueONNXMatchingOptions, **kwargs) -> None`
```

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

### SequentialPairingOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration options for sequential image pairing.

```APIDOC
## SequentialPairingOptions

### Description
Configuration class for sequential image matching strategies.

### Properties
- **overlap** (int) - Number of overlapping image pairs. Default: 10.
- **quadratic_overlap** (bool) - Whether to match images against quadratic neighbors. Default: True.
- **expand_rig_images** (bool) - Whether to match against neighboring rig frames. Default: True.
- **loop_detection** (bool) - Whether to enable loop detection. Default: False.
- **loop_detection_period** (int) - Frequency of loop detection. Default: 10.
- **loop_detection_num_images** (int) - Number of images to retrieve in loop detection. Default: 50.
```

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

### AlikedMatchingOptions Initialization

Source: https://colmap.github.io/pycolmap/pycolmap.html

Options for configuring Aliked matching. Supports initialization with default values, a dictionary, or keyword arguments.

```python
class pycolmap.AlikedMatchingOptions(_* args_, _** kwargs_)
    
Overloaded function.
  1. `__init__(self: pycolmap.AlikedMatchingOptions) -> None`
  2. `__init__(self: pycolmap.AlikedMatchingOptions, arg0: dict) -> None`
  3. `__init__(self: pycolmap.AlikedMatchingOptions, **kwargs) -> None`
```

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

### Initialize DelaunayMeshingOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Overloaded constructors for configuring Delaunay meshing parameters.

```python
1. `__init__(self: pycolmap.DelaunayMeshingOptions) → None`
  2. `__init__(self: pycolmap.DelaunayMeshingOptions, arg0: dict) → None`
```

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

### Perform Feature Matching

Source: https://colmap.github.io/pycolmap/pycolmap.html

Methods for standard and guided feature matching between two images.

```python
match( 
    _self : pycolmap.FeatureMatcher_,     _keypoints1 : pycolmap.FeatureKeypoints_,     _descriptors1 : pycolmap.FeatureDescriptors_,     _keypoints2 : pycolmap.FeatureKeypoints_,     _descriptors2 : pycolmap.FeatureDescriptors_, ) → Annotated[numpy.typing.NDArray[numpy.uint32], '[m, 2]']
```

```python
match_guided( 
    _self : pycolmap.FeatureMatcher_,     _max_error : SupportsFloat_,     _keypoints1 : pycolmap.FeatureKeypoints_,     _descriptors1 : pycolmap.FeatureDescriptors_,     _camera1 : pycolmap.Camera_,     _keypoints2 : pycolmap.FeatureKeypoints_,     _descriptors2 : pycolmap.FeatureDescriptors_,     _camera2 : pycolmap.Camera_,     _two_view_geometry : pycolmap.TwoViewGeometry_, ) → None
```

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

### Ceres Solver Options

Source: https://colmap.github.io/pycolmap/pycolmap.html

Options for the Ceres solver. Using this member requires having PyCeres installed.

```APIDOC
## Ceres Solver Options

### Description
Options for the Ceres solver. Using this member requires having PyCeres installed.

### Properties
- **minimizer_type** (MinimizerType) - Default: MinimizerType.TRUST_REGION - Type of minimizer to use.
- **line_search_direction_type** (LineSearchDirectionType) - Default: LineSearchDirectionType.LBFGS - Type of line search direction.
- **line_search_type** (LineSearchType) - Default: LineSearchType.WOLFE - Type of line search.
- **nonlinear_conjugate_gradient_type** (NonlinearConjugateGradientType) - Default: NonlinearConjugateGradientType.FLETCHER_REEVES - Type of nonlinear conjugate gradient.
- **max_lbfgs_rank** (int) - Default: 20 - Maximum rank for L-BFGS.
- **use_approximate_eigenvalue_bfgs_scaling** (bool) - Default: False - Whether to use approximate eigenvalue scaling for BFGS.
- **line_search_interpolation_type** (LineSearchInterpolationType) - Default: LineSearchInterpolationType.CUBIC - Type of line search interpolation.
- **min_line_search_step_size** (float) - Default: 1e-09 - Minimum line search step size.
- **line_search_sufficient_function_decrease** (float) - Default: 0.0001 - Sufficient function decrease for line search.
- **max_line_search_step_contraction** (float) - Default: 0.001 - Maximum step contraction for line search.
- **min_line_search_step_contraction** (float) - Default: 0.6 - Minimum step contraction for line search.
- **max_num_line_search_step_size_iterations** (int) - Default: 20 - Maximum iterations for line search step size.
- **max_num_line_search_direction_restarts** (int) - Default: 5 - Maximum restarts for line search direction.
- **line_search_sufficient_curvature_decrease** (float) - Default: 0.9 - Sufficient curvature decrease for line search.
- **max_line_search_step_expansion** (float) - Default: 10.0 - Maximum step expansion for line search.
- **trust_region_strategy_type** (TrustRegionStrategyType) - Default: TrustRegionStrategyType.LEVENBERG_MARQUARDT - Type of trust region strategy.
- **dogleg_type** (DoglegType) - Default: DoglegType.TRADITIONAL_DOGLEG - Type of dogleg.
- **use_nonmonotonic_steps** (bool) - Default: False - Whether to use non-monotonic steps.
- **max_consecutive_nonmonotonic_steps** (int) - Default: 10 - Maximum consecutive non-monotonic steps.
- **max_num_iterations** (int) - Default: 100 - Maximum number of iterations.
- **max_solver_time_in_seconds** (float) - Default: 1000000000.0 - Maximum solver time in seconds.
- **num_threads** (int) - Default: -1 - Number of threads to use (-1 for auto).
- **initial_trust_region_radius** (float) - Default: 10000.0 - Initial trust region radius.
- **max_trust_region_radius** (float) - Default: 1e+16 - Maximum trust region radius.
- **min_trust_region_radius** (float) - Default: 1e-32 - Minimum trust region radius.
- **min_relative_decrease** (float) - Default: 0.001 - Minimum relative decrease.
- **min_lm_diagonal** (float) - Default: 1e-06 - Minimum Levenberg-Marquardt diagonal.
- **max_lm_diagonal** (float) - Default: 1e+32 - Maximum Levenberg-Marquardt diagonal.
- **max_num_consecutive_invalid_steps** (int) - Default: 10 - Maximum consecutive invalid steps.
- **function_tolerance** (float) - Default: 0.0 - Function tolerance.
- **gradient_tolerance** (float) - Default: 0.0001 - Gradient tolerance.
- **parameter_tolerance** (float) - Default: 0.0 - Parameter tolerance.
- **linear_solver_type** (LinearSolverType) - Default: LinearSolverType.SPARSE_NORMAL_CHOLESKY - Type of linear solver.
- **preconditioner_type** (PreconditionerType) - Default: PreconditionerType.JACOBI - Type of preconditioner.
- **visibility_clustering_type** (VisibilityClusteringType) - Default: VisibilityClusteringType.CANONICAL_VIEWS - Type of visibility clustering.
- **dense_linear_algebra_library_type** (DenseLinearAlgebraLibraryType) - Default: DenseLinearAlgebraLibraryType.EIGEN - Type of dense linear algebra library.
- **sparse_linear_algebra_library_type** (SparseLinearAlgebraLibraryType) - Default: SparseLinearAlgebraLibraryType.SUITE_SPARSE - Type of sparse linear algebra library.
- **use_explicit_schur_complement** (bool) - Default: False - Whether to use explicit Schur complement.
- **dynamic_sparsity** (bool) - Default: False - Whether to use dynamic sparsity.
- **use_inner_iterations** (bool) - Default: False - Whether to use inner iterations.
- **inner_iteration_tolerance** (float) - Default: 0.001 - Tolerance for inner iterations.
- **min_linear_solver_iterations** (int) - Default: 0 - Minimum linear solver iterations.
- **max_linear_solver_iterations** (int) - Default: 200 - Maximum linear solver iterations.
- **eta** (float) - Default: 0.1 - Eta parameter for trust region.
- **jacobi_scaling** (bool) - Default: True - Whether to use Jacobi scaling.
- **logging_type** (LoggingType) - Default: LoggingType.SILENT - Type of logging.
- **minimizer_progress_to_stdout** (bool) - Default: False - Whether to print minimizer progress to stdout.
- **trust_region_problem_dump_directory** (str) - Default: '/tmp' - Directory to dump trust region problems.
- **trust_region_problem_dump_format_type** (DumpFormatType) - Default: DumpFormatType.TEXTFILE - Format type for dumping trust region problems.
- **check_gradients** (bool) - Default: False - Whether to check gradients.
- **gradient_check_relative_precision** (float) - Default: 1e-08 - Relative precision for gradient check.
- **gradient_check_numeric_derivative_relative_step_size** (float) - Default: 1e-06 - Relative step size for numeric derivative in gradient check.
- **update_state_every_iteration** (bool) - Default: False - Whether to update state every iteration.

### Methods
- **check()**: Returns true if the options are valid.
- **summary(write_type=False)**: Returns a summary string of the options.
- **mergedict(**kwargs**)**: Merges options from a dictionary.
- **todict(recursive=True)**: Converts the options to a dictionary.
```

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

### Initialize PoissonMeshingOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Overloaded constructors for configuring Poisson meshing parameters.

```python
1. `__init__(self: pycolmap.PoissonMeshingOptions) → None`
  2. `__init__(self: pycolmap.PoissonMeshingOptions, arg0: dict) → None`
  3. `__init__(self: pycolmap.PoissonMeshingOptions, **kwargs) → None`
```

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

### Initialize StereoFusionOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Overloaded constructors for configuring stereo fusion parameters.

```python
1. `__init__(self: pycolmap.StereoFusionOptions) → None`
  2. `__init__(self: pycolmap.StereoFusionOptions, arg0: dict) → None`
  3. `__init__(self: pycolmap.StereoFusionOptions, **kwargs) → None`
```

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

### GET /compare_reconstructions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Compares two reconstructions based on reprojection error and projection center error.

```APIDOC
## GET /compare_reconstructions

### Description
Compares two reconstruction objects and returns a dictionary of metrics.

### Method
GET

### Endpoint
/compare_reconstructions

### Parameters
#### Query Parameters
- **reconstruction1** (pycolmap.Reconstruction) - Required - First reconstruction.
- **reconstruction2** (pycolmap.Reconstruction) - Required - Second reconstruction.
- **alignment_error** (str) - Optional - Type of error, default 'reprojection'.
- **min_inlier_observations** (SupportsFloat) - Optional - Default 0.3.
- **max_reproj_error** (SupportsFloat) - Optional - Default 8.0.
- **max_proj_center_error** (SupportsFloat) - Optional - Default 0.1.

### Response
#### Success Response (200)
- **result** (dict | None) - Dictionary containing comparison metrics.
```

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

### SiftMatchingOptions Initialization

Source: https://colmap.github.io/pycolmap/pycolmap.html

Options for configuring SIFT matching. Supports initialization with default values, a dictionary, or keyword arguments.

```python
class pycolmap.SiftMatchingOptions(_* args_, _** kwargs_)
    
Overloaded function.
  1. `__init__(self: pycolmap.SiftMatchingOptions) -> None`
  2. `__init__(self: pycolmap.SiftMatchingOptions, arg0: dict) -> None`
  3. `__init__(self: pycolmap.SiftMatchingOptions, **kwargs) -> None`
```

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

### Build PyCOLMAP from source on Windows

Source: https://colmap.github.io/pycolmap/index.html

Build PyCOLMAP from source on Windows after installing COLMAP via VCPKG. This command specifies the CMake toolchain file and target triplet for VCPKG.

```powershell
python -m pip install . `
    --cmake.define.CMAKE_TOOLCHAIN_FILE="$VCPKG_ROOT/scripts/buildsystems/vcpkg.cmake" `
    --cmake.define.VCPKG_TARGET_TRIPLET="x64-windows"
```

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

### Create FeatureMatcher

Source: https://colmap.github.io/pycolmap/pycolmap.html

Static method to instantiate a feature matcher with specific options and device.

```python
static create( 
    _options : pycolmap.FeatureMatchingOptions | None = None_,     _device : pycolmap.Device = Device.auto_, ) → pycolmap.FeatureMatcher
```

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

### Get Other Parameters Covariance

Source: https://colmap.github.io/pycolmap/pycolmap.html

Calculates the tangent space covariance for any variable parameter block. Returns null if the parameter block is not a variable. Rows/columns are omitted if dimensions are kept constant.

```python
get_other_params_cov(
    _self : pycolmap.BACovariance_,
    _param : Annotated[numpy.typing.ArrayLike, numpy.float64]_,
) → Annotated[numpy.typing.NDArray[numpy.float64], '[m, n]'] | None
```

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

### Get Camera Covariance from World

Source: https://colmap.github.io/pycolmap/pycolmap.html

Calculates the covariance for a camera's pose relative to the world. Returns null if the image is not a variable. Covariance is in tangent space order [rotation, translation].

```python
get_cam_cov_from_world(
    _self : pycolmap.BACovariance_,
    _image_id : SupportsInt_,
) → Annotated[numpy.typing.NDArray[numpy.float64], '[m, n]'] | None
```

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

### Core Components and Options

Source: https://colmap.github.io/pycolmap/index.html

Overview of key classes and options for configuring COLMAP pipelines.

```APIDOC
## Core Components and Options

### Description
This section lists the primary classes and option structures used for configuring and managing COLMAP operations.

### Classes
- `IncrementalTriangulatorOptions`
- `IncrementalTriangulator`
- `ImageSelectionMethod`
- `IncrementalMapperOptions`
- `LocalBundleAdjustmentReport`
- `IncrementalMapper`
- `IncrementalPipelineOptions`
- `IncrementalPipelineCallback`
- `IncrementalPipelineStatus`
- `IncrementalPipeline`
- `DepthMap`
- `NormalMap`
- `MVSModel`
- `CameraMode`
- `ImageReaderOptions`
- `FileCopyType`
- `PairGenerator`
- `ExhaustivePairGenerator`
- `VocabTreePairGenerator`
- `SequentialPairGenerator`
- `SpatialPairGenerator`
- `ImportedPairGenerator`

### Cost Functions
- `ReprojErrorCost()`
- `RigReprojErrorCost()`
- `SampsonErrorCost()`
- `AbsolutePosePriorCost()`
- `AbsolutePosePositionPriorCost()`
- `RelativePosePriorCost()`
- `Point3DAlignmentCost()`
```

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

### Get Relative Pose Covariance Between Cameras

Source: https://colmap.github.io/pycolmap/pycolmap.html

Calculates the relative pose covariance between two cameras. Returns null if dimensions are kept constant for either pose, requiring custom logic for such cases.

```python
get_cam2_cov_from_cam1(
    _self : pycolmap.BACovariance_,
    _image_id1 : SupportsInt_,
    _cam1_from_world : pycolmap.Rigid3d_,
    _image_id2 : SupportsInt_,
    _cam2_from_world : pycolmap.Rigid3d_,
) → Annotated[numpy.typing.NDArray[numpy.float64], '[m, n]'] | None
```

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

### Initialize Incremental Pipeline

Source: https://colmap.github.io/pycolmap/pycolmap.html

Constructs an IncrementalPipeline object. Use the database cache for faster initialization if available.

```python
IncrementalPipeline(options: pycolmap.IncrementalPipelineOptions, database: pycolmap.Database, reconstruction_manager: pycolmap.ReconstructionManager) -> None
IncrementalPipeline(options: pycolmap.IncrementalPipelineOptions, database_cache: pycolmap.DatabaseCache, reconstruction_manager: pycolmap.ReconstructionManager) -> None
```

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

### Get Cross Camera Covariance from World

Source: https://colmap.github.io/pycolmap/pycolmap.html

Calculates the cross-covariance between two cameras' poses relative to the world. Returns null if either image is not a variable. Covariance is in tangent space order [rotation, translation].

```python
get_cam_cross_cov_from_world(
    _self : pycolmap.BACovariance_,
    _image_id1 : SupportsInt_,
    _image_id2 : SupportsInt_,
) → Annotated[numpy.typing.NDArray[numpy.float64], '[m, n]'] | None
```

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

### Initialize FeatureMatchingOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Overloaded constructors for creating a FeatureMatchingOptions object.

```python
__init__(self: pycolmap.FeatureMatchingOptions, type: pycolmap.FeatureMatcherType = <FeatureMatcherType.SIFT_BRUTEFORCE: 0>) → None
```

```python
__init__(self: pycolmap.FeatureMatchingOptions, arg0: dict) → None
```

```python
__init__(self: pycolmap.FeatureMatchingOptions, **kwargs) → None
```

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

### LightGlueONNXMatchingOptions Methods

Source: https://colmap.github.io/pycolmap/pycolmap.html

Utility methods for LightGlueONNXMatchingOptions, including checking validity, generating a summary string, merging dictionaries, and converting to a dictionary.

```python
check(_self : pycolmap.LightGlueONNXMatchingOptions_) -> bool
```

```python
summary(_self : pycolmap.LightGlueONNXMatchingOptions_, _write_type : bool = False_) -> str
```

```python
mergedict(_self : object_, _kwargs : dict_) -> None
```

```python
todict(_self : pycolmap.LightGlueONNXMatchingOptions_, _recursive : bool = True_) -> dict
```

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

### BruteForceONNXMatchingOptions Initialization

Source: https://colmap.github.io/pycolmap/pycolmap.html

Options for configuring brute-force matching using ONNX models. Supports initialization with default values, a dictionary, or keyword arguments.

```python
class pycolmap.BruteForceONNXMatchingOptions(_* args_, _** kwargs_)
    
Overloaded function.
  1. `__init__(self: pycolmap.BruteForceONNXMatchingOptions) -> None`
  2. `__init__(self: pycolmap.BruteForceONNXMatchingOptions, arg0: dict) -> None`
  3. `__init__(self: pycolmap.BruteForceONNXMatchingOptions, **kwargs) -> None`
```

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

### Manage Reconstruction Lifecycle

Source: https://colmap.github.io/pycolmap/pycolmap.html

Methods to prepare and clean up the mapper for a reconstruction session.

```python
begin_reconstruction( 
    _self : pycolmap.IncrementalMapper_,     _reconstruction : pycolmap.Reconstruction_, ) → None
```

```python
end_reconstruction(_self : pycolmap.IncrementalMapper_, _discard : bool_) → None
```

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

### IncrementalPipelineOptions Methods

Source: https://colmap.github.io/pycolmap/pycolmap.html

Methods for retrieving configuration options and checking pipeline status.

```APIDOC
## IncrementalPipelineOptions Methods

### Description
Methods for retrieving configuration options and checking pipeline status.

### Methods
- **get_mapper** (_self: pycolmap.IncrementalPipelineOptions_) -> pycolmap.IncrementalMapperOptions: Get mapper options with shared settings applied.
- **get_triangulation** (_self: pycolmap.IncrementalPipelineOptions_) -> pycolmap.IncrementalTriangulatorOptions: Get triangulation options with shared settings applied.
- **get_local_bundle_adjustment** (_self: pycolmap.IncrementalPipelineOptions_) -> pycolmap.BundleAdjustmentOptions: Get local bundle adjustment options.
- **get_global_bundle_adjustment** (_self: pycolmap.IncrementalPipelineOptions_) -> pycolmap.BundleAdjustmentOptions: Get global bundle adjustment options.
- **is_initial_pair_provided** (_self: pycolmap.IncrementalPipelineOptions_) -> bool: Check whether both initial image identifiers are provided.
- **check** (_self: pycolmap.IncrementalPipelineOptions_) -> bool: 
- **summary** (_self: pycolmap.IncrementalPipelineOptions_, _write_type: bool = False_) -> str: 
- **mergedict** (_self: object_, _kwargs: dict_) -> None: 
- **todict** (_self: pycolmap.IncrementalPipelineOptions_, _recursive: bool = True_) -> dict: 
```

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

### VocabTreePairingOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration options for vocabulary tree-based image pairing.

```APIDOC
## VocabTreePairingOptions

### Description
Configuration class for vocabulary tree pairing settings.

### Properties
- **num_images_after_verification** (int) - How many images to return after spatial verification. Default: 0.
- **max_num_features** (int) - Maximum number of features to use for indexing. Default: -1.
- **vocab_tree_path** (PosixPath) - Path to the vocabulary tree. Default: .
- **match_list_path** (PosixPath) - Optional path to file with specific image names to match. Default: .
- **num_threads** (int) - Number of threads. Default: -1.
```

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

### Image Class Overview

Source: https://colmap.github.io/pycolmap/pycolmap.html

Provides an overview of the pycolmap.Image class, its constructors, and key properties.

```APIDOC
## Class pycolmap.Image

### Description
Represents an image within the COLMAP structure, holding information about its keypoints, camera, and pose.

### Constructors
Overloaded function with multiple signatures:
1. `__init__(self: pycolmap.Image) -> None`
2. `__init__(self: pycolmap.Image, name: str = '', points2D: pycolmap.Point2DList = Point2DList(), camera_id: typing.SupportsInt = pycolmap.INVALID_CAMERA_ID, image_id: typing.SupportsInt = pycolmap.INVALID_IMAGE_ID) -> None`
3. `__init__(self: pycolmap.Image, name: str = '', keypoints: typing.Annotated[numpy.typing.ArrayLike, numpy.float64, "[m, 2]"] = array([], shape=(0, 2), dtype=float64), camera_id: typing.SupportsInt = pycolmap.INVALID_CAMERA_ID, image_id: typing.SupportsInt = pycolmap.INVALID_IMAGE_ID) -> None`
4. `__init__(self: pycolmap.Image, arg0: dict) -> None`
5. `__init__(self: pycolmap.Image, **kwargs) -> None`

### Properties
- **image_id** (int) - Unique identifier of the image. (default: 4294967295)
- **camera_id** (int) - Unique identifier of the camera. (default: 4294967295)
- **frame_id** (int) - Unique identifier of the frame. (default: 4294967295)
- **data_id** (data_t) - Unique identifier of the data. (default: data_t(sensor_id=sensor_t(type=SensorType.CAMERA, id=4294967295), id=4294967295))
- **camera** (NoneType) - The associated camera object. (default: None)
- **frame** (NoneType) - The associated frame object. (default: None)
- **name** (str) - Name of the image. (default: )
- **has_pose** (bool) - Whether the image has a valid pose. (default: False)
- **points2D** (Point2DList) - Array of Points2D (=keypoints). (default: Point2DList[])
- **num_points3D** (int) - Get the number of triangulations, i.e. the number of points that are part of a 3D point track. (default: 0)
```

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

### Initialize Reconstruction

Source: https://colmap.github.io/pycolmap/pycolmap.html

Initializes a new reconstruction by finding and registering an initial image pair. Requires a mapper, mapper options, and a reconstruction object.

```python
initialize_reconstruction( _self : pycolmap.IncrementalPipeline_, _mapper : pycolmap.IncrementalMapper_, _mapper_options : pycolmap.IncrementalMapperOptions_, _reconstruction : pycolmap.Reconstruction_, ) -> pycolmap.IncrementalPipelineStatus
```

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

### PyCOLMAP Rig and Dataset Synthesis

Source: https://colmap.github.io/pycolmap/pycolmap.html

This section covers functionalities for handling camera rigs and synthesizing datasets for testing and development.

```APIDOC
## PyCOLMAP Rig and Dataset Synthesis

### Description
This section details the components for defining and utilizing camera rigs, as well as synthesizing datasets and noise for experimental purposes.

### Rig Handling

- **`Rig`**: Represents a collection of calibrated cameras mounted on a rigid structure.
- **`RigMap`**: A map from rig ID to `Rig` objects.
- **`read_rig_config(path)`**: Reads rig configuration from a file.
- **`apply_rig_config(rig_config, image)`**: Applies rig configuration to an image.

### Dataset Synthesis

- **`SyntheticDatasetMatchConfig`**: Configuration for matching in synthetic datasets.
- **`SyntheticDatasetOptions`**: General options for synthesizing a dataset.
- **`synthesize_dataset(options)`**: Generates a synthetic dataset based on provided options.

### Noise Synthesis

- **`SyntheticNoiseOptions`**: Options for synthesizing noise.
- **`synthesize_noise(options)`**: Adds synthetic noise to data.

### Image Synthesis

- **`SyntheticImageOptions`**: Options for synthesizing images.
- **`synthesize_images(options)`**: Generates synthetic images.
```

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

### AlikedMatchingOptions Methods

Source: https://colmap.github.io/pycolmap/pycolmap.html

Utility methods for AlikedMatchingOptions, including checking validity, generating a summary string, merging dictionaries, and converting to a dictionary.

```python
check(_self : pycolmap.AlikedMatchingOptions_) -> bool
```

```python
summary(_self : pycolmap.AlikedMatchingOptions_, _write_type : bool = False_) -> str
```

```python
mergedict(_self : object_, _kwargs : dict_) -> None
```

```python
todict(_self : pycolmap.AlikedMatchingOptions_, _recursive : bool = True_) -> dict
```

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

### ImportedPairingOptions

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration options for importing pre-defined image matches.

```APIDOC
## ImportedPairingOptions

### Description
Configuration class for importing existing match lists.

### Properties
- **block_size** (int) - Number of image pairs to match in one batch. Default: 1225.
- **match_list_path** (PosixPath) - Path to the file with the matches. Default: .
```

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

### Initialize Reconstruction Geometry

Source: https://colmap.github.io/pycolmap/pycolmap.html

Methods for finding and registering initial image pairs to seed the reconstruction.

```python
find_initial_image_pair( 
    _self : pycolmap.IncrementalMapper_,     _options : pycolmap.IncrementalMapperOptions_,     _image_id1 : SupportsInt_,     _image_id2 : SupportsInt_, ) → tuple[tuple[int, int], pycolmap.Rigid3d] | None
```

```python
estimate_initial_two_view_geometry( 
    _self : pycolmap.IncrementalMapper_,     _options : pycolmap.IncrementalMapperOptions_,     _image_id1 : SupportsInt_,     _image_id2 : SupportsInt_, ) → pycolmap.Rigid3d | None
```

```python
register_initial_image_pair( 
    _self : pycolmap.IncrementalMapper_,     _options : pycolmap.IncrementalMapperOptions_,     _two_view_geometry : SupportsInt_,     _image_id1 : SupportsInt_,     _image_id2 : pycolmap.Rigid3d_, ) → None
```

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

### FeatureMatcher.create

Source: https://colmap.github.io/pycolmap/pycolmap.html

Creates a new feature matcher instance based on the provided options and device.

```APIDOC
## static create

### Description
Creates a new feature matcher instance.

### Parameters
#### Query Parameters
- **options** (pycolmap.FeatureMatchingOptions | None) - Optional - Configuration options for the matcher.
- **device** (pycolmap.Device) - Optional - The device to use for matching (default: Device.auto).

### Response
- **returns** (pycolmap.FeatureMatcher) - A new instance of the feature matcher.
```

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

### Data Import and Preprocessing Functions

Source: https://colmap.github.io/pycolmap/index.html

Functions for importing images and performing initial camera and image processing.

```APIDOC
## Data Import and Preprocessing Functions

### Description
These functions handle the initial steps of importing image data and inferring camera parameters.

### Functions
- `import_images()`: Imports images into the COLMAP project.
- `infer_camera_from_image()`: Infers camera parameters from an image.
- `undistort_images()`: Undistorts images based on inferred or provided camera parameters.
```

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

### Run Patch-Match Stereo

Source: https://colmap.github.io/pycolmap/pycolmap.html

Executes the Patch-Match-Stereo algorithm, which requires CUDA support.

```python
pycolmap.patch_match_stereo( 
    _workspace_path : os.PathLike | str | bytes_,     _workspace_format : str = 'COLMAP'_,     _pmvs_option_name : str = 'option-all'_,     _options : pycolmap.PatchMatchOptions = PatchMatchOptions()_,     _config_path : os.PathLike | str | bytes = ''_, ) → None
```

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

### Initialize TrackElement

Source: https://colmap.github.io/pycolmap/pycolmap.html

Overloaded constructors for creating a TrackElement object.

```python
__init__(self: pycolmap.TrackElement) → None
```

```python
__init__(self: pycolmap.TrackElement, image_id: typing.SupportsInt, point2D_idx: typing.SupportsInt) → None
```

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

### AlikedMatchingOptions Properties

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration properties for Aliked matching, including options for brute-force and LightGlue sub-matchers.

```python
property brute_force
    
Brute-force matching options. (BruteForceONNXMatchingOptions, default: BruteForceONNXMatchingOptions(min_cossim=0.85, max_ratio=1.0, cross_check=True, model_path=’https://github.com/colmap/colmap/releases/download/3.13.0/bruteforce-matcher.onnx;bruteforce-matcher.onnx;3c1282f96d83f5ffc861a873298d08bbe5219f59af59223f5ceab5c41a182a47’)) 
```

```python
property lightglue
    
LightGlue matching options. (LightGlueONNXMatchingOptions, default: LightGlueONNXMatchingOptions(min_score=0.1, model_path=’https://github.com/colmap/colmap/releases/download/3.13.0/aliked-lightglue.onnx;aliked-lightglue.onnx;b9a5de7204648b18a8cf5dcac819f9d30de1a5961ef03756803c8b86c2dceb8d’)) 
```

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

### Rig Configuration

Source: https://colmap.github.io/pycolmap/pycolmap.html

Classes and methods for defining and managing rig configurations, including camera and sensor properties.

```APIDOC
## Class: RigConfigCamera

### Description
Represents camera configuration within a rig.

### Properties
- **ref_sensor** (bool, default: False) - Indicates if this camera is a reference sensor.
- **image_prefix** (str, default: '') - Prefix for images associated with this camera.
- **cam_from_rig** (NoneType, default: None) - Transformation from rig to camera coordinates.
- **camera** (NoneType, default: None) - The camera intrinsic parameters.

### Methods

#### GET /rig_config_camera/summary

##### Description
Get a summary string of the rig camera configuration.

##### Parameters
- **write_type** (bool, default: False) - Whether to include type information in the summary.

##### Response
- **summary** (str)

#### POST /rig_config_camera/mergedict

##### Description
Merge a dictionary of keyword arguments into the rig camera configuration.

##### Parameters
- **kwargs** (dict) - The dictionary of parameters to merge.

#### GET /rig_config_camera/todict

##### Description
Convert the rig camera configuration to a dictionary.

##### Parameters
- **recursive** (bool, default: True) - Whether to recursively convert nested objects.

##### Response
- **dictionary** (dict)

## Class: RigConfig

### Description
Represents a configuration for a camera rig.
```

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

### Initialize LocalBundleAdjustmentReport

Source: https://colmap.github.io/pycolmap/pycolmap.html

Overloaded constructors for the LocalBundleAdjustmentReport class.

```python
1. __init__(self: pycolmap.LocalBundleAdjustmentReport) → None
2. __init__(self: pycolmap.LocalBundleAdjustmentReport, arg0: dict) → None
3. __init__(self: pycolmap.LocalBundleAdjustmentReport, **kwargs) → None
```

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

### IncrementalPipeline Constructor

Source: https://colmap.github.io/pycolmap/pycolmap.html

Initializes the IncrementalPipeline class. It can be constructed with either a Database or a DatabaseCache.

```python
IncrementalPipeline(_* args_, _** kwargs_)
__init__(self: pycolmap.IncrementalPipeline, options: pycolmap.IncrementalPipelineOptions, database: pycolmap.Database, reconstruction_manager: pycolmap.ReconstructionManager) -> None
__init__(self: pycolmap.IncrementalPipeline, options: pycolmap.IncrementalPipelineOptions, database_cache: pycolmap.DatabaseCache, reconstruction_manager: pycolmap.ReconstructionManager) -> None
```

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

### LightGlueONNXMatchingOptions Properties

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration properties for ONNX LightGlue matching, including the minimum score threshold and the path to the ONNX model file.

```python
property min_score
    
Minimum match score threshold. (float, default: 0.1) 
```

```python
property model_path
    
Path to the LightGlue ONNX model file. (str, default: ) 
```

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

### BundleAdjustmentConfig Methods

Source: https://colmap.github.io/pycolmap/pycolmap.html

Methods for configuring the bundle adjustment process, including managing images and camera intrinsics.

```APIDOC
## BundleAdjustmentConfig Methods

### Description
Methods to configure the bundle adjustment process by adding/removing images and setting camera intrinsic constraints.

### Methods
- **add_image(image_id: SupportsInt) -> None**: Adds an image to the bundle adjustment configuration.
- **has_image(image_id: SupportsInt) -> bool**: Checks if an image is included in the configuration.
- **remove_image(image_id: SupportsInt) -> None**: Removes an image from the configuration.
- **set_constant_cam_intrinsics(camera_id: SupportsInt) -> None**: Sets camera intrinsics as constant.
- **set_variable_cam_intrinsics(camera_id: SupportsInt) -> None**: Sets camera intrinsics as variable.
```

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

### PyCOLMAP AlikedExtractionOptions Class

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration options for Aliked feature extraction.

```APIDOC
## PyCOLMAP AlikedExtractionOptions Class

### Description
Configuration options for Aliked feature extraction.

### Methods
- `__init__(self: pycolmap.AlikedExtractionOptions, *args, **kwargs) → None`
```

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

### Rig Configuration API

Source: https://colmap.github.io/pycolmap/pycolmap.html

Functions for reading and applying rig configurations to a database.

```APIDOC
## pycolmap.read_rig_config

### Description
Read the rig configuration from a .json file.

### Parameters
- **path** (os.PathLike | str | bytes) - Required - The path to the rig configuration file.

### Response
- **list[pycolmap.RigConfig]** - A list of rig configurations.

## pycolmap.apply_rig_config

### Description
Applies the given rig configuration to the database and optionally derives camera rig extrinsics and intrinsics from the reconstruction.

### Parameters
- **configs** (collections.abc.Sequence[pycolmap.RigConfig]) - Required - The rig configurations to apply.
- **database** (pycolmap.Database) - Required - The database to update.
- **reconstruction** (pycolmap.Reconstruction) - Optional - The reconstruction to update.
```

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

### Absolute Pose Refinement Options

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configure options for non-linear absolute pose refinement. Includes iteration limits, tolerances, and prior settings.

```python
class pycolmap.AbsolutePoseRefinementOptions(_* args_, _** kwargs_){
    
    Overloaded function.
      1. `__init__(self: pycolmap.AbsolutePoseRefinementOptions) -> None`
      2. `__init__(self: pycolmap.AbsolutePoseRefinementOptions, arg0: dict) -> None`
      3. `__init__(self: pycolmap.AbsolutePoseRefinementOptions, **kwargs) -> None`



    property gradient_tolerance
        
        (float, default: 1.0)

    property max_num_iterations
        
        (int, default: 100)

    property loss_function_scale
        
        (float, default: 1.0)

    property refine_focal_length
        
        (bool, default: False)

    property refine_extra_params
        
        (bool, default: False)

    property print_summary
        
        (bool, default: False)

    property use_position_prior
        
        (bool, default: False)

    property position_prior_in_world
        
        (ndarray, default: [0. 0. 0.])

    property position_prior_covariance
        
        (ndarray, default: [[1. 0. 0.] [0. 1. 0.] [0. 0. 1.]])

    summary(_self : pycolmap.AbsolutePoseRefinementOptions_, _write_type : bool = False_) -> str
         

    mergedict(_self : object_, _kwargs : dict_) -> None
         

    todict(_self : pycolmap.AbsolutePoseRefinementOptions_, _recursive : bool = True_) -> dict
         
}
```

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

### Run Incremental Mapping Pipeline

Source: https://colmap.github.io/pycolmap/pycolmap.html

Executes the full incremental mapping pipeline. This method should be called after the pipeline has been configured.

```python
run(_self : pycolmap.IncrementalPipeline_) -> None
```

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

### pycolmap.logging

Source: https://colmap.github.io/pycolmap/pycolmap.html

Provides logging capabilities with different levels and configuration options.

```APIDOC
## pycolmap.logging

### Description
This module provides functionalities for logging messages within the pycolmap library. It supports various log levels and allows configuration of log destinations and verbosity.

### Classes
- `Level`: An enumeration for log levels (INFO, WARNING, ERROR, FATAL).

### Level Members
- `INFO` (0)
- `WARNING` (1)
- `ERROR` (2)
- `FATAL` (3)

### Global Logging Settings
- `minloglevel` (int): Minimum log level to report.
- `stderrthreshold` (int): Threshold for logging to stderr.
- `log_dir` (str): Directory for log files.
- `logtostderr` (bool): Whether to log to stderr.
- `alsologtostderr` (bool): Whether to also log to stderr.
- `logtostdout` (bool): Whether to log to stdout.
- `colorlogtostdout` (bool): Whether to use color for stdout logging.
- `colorlogtostderr` (bool): Whether to use color for stderr logging.
- `verbose_level` (int): Verbosity level for detailed logging.

### Static Methods
- `set_log_destination(level: pycolmap.logging.Level, path: os.PathLike | str | bytes) -> None`: Sets the destination for log files.
- `verbose(level: SupportsInt, message: str) -> None`: Logs a message at a specified verbose level.
- `info(message: str) -> None`: Logs an informational message.
- `warning(message: str) -> None`: Logs a warning message.
- `error(message: str) -> None`: Logs an error message.
- `fatal(message: str) -> None`: Logs a fatal error message.
```

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

### Poisson Meshing Options

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configuration options for Poisson surface reconstruction.

```APIDOC
## POISSON MESHING OPTIONS

### Description
Configuration options for the Poisson surface reconstruction algorithm, used to generate a mesh from a point cloud.

### Class
`pycolmap.PoissonMeshingOptions`

### Properties
- `point_weight` (float, default: 1.0) - Importance of point samples in the screened Poisson equation formulation.
- `depth` (int, default: 13) - Maximum depth of the tree for surface reconstruction. Corresponds to a maximum voxel grid resolution of 2^depth x 2^depth x 2^depth.
- `color` (bool, default: True) - If True, assumes input has colors and extrapolates them to the mesh vertices. The float value specifies the relative importance of finer color estimates.
- `trim` (float, default: 10.0) - Value for mesh trimming. Subsets with signal values less than this are discarded.
- `num_threads` (int, default: -1) - The number of threads used for the Poisson reconstruction.

### Methods
- `check(self: pycolmap.PoissonMeshingOptions) -> bool`
- `summary(self: pycolmap.PoissonMeshingOptions, _write_type: bool = False) -> str`
- `mergedict(self: object, **kwargs: dict) -> None`
- `todict(self: pycolmap.PoissonMeshingOptions, _recursive: bool = True) -> dict`
```

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

### Absolute Pose Estimation Options

Source: https://colmap.github.io/pycolmap/pycolmap.html

Configure options for absolute pose estimation. Includes RANSAC parameters and focal length estimation.

```python
class pycolmap.AbsolutePoseEstimationOptions(_* args_, _** kwargs_){
    
    Overloaded function.
      1. `__init__(self: pycolmap.AbsolutePoseEstimationOptions) -> None`
      2. `__init__(self: pycolmap.AbsolutePoseEstimationOptions, arg0: dict) -> None`
      3. `__init__(self: pycolmap.AbsolutePoseEstimationOptions, **kwargs) -> None`



    property estimate_focal_length
        
        (bool, default: False)

    property ransac
        
        (RANSACOptions, default: RANSACOptions(max_error=12.0, min_inlier_ratio=0.1, confidence=0.99999, dyn_num_trials_multiplier=3.0, min_num_trials=100, max_num_trials=10000, random_seed=-1, num_threads=1))

    summary(_self : pycolmap.AbsolutePoseEstimationOptions_, _write_type : bool = False_) -> str
         

    mergedict(_self : object_, _kwargs : dict_) -> None
         

    todict(_self : pycolmap.AbsolutePoseEstimationOptions_, _recursive : bool = True_) -> dict
         
}
```

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

### Camera Class API

Source: https://colmap.github.io/pycolmap/pycolmap.html

Methods and properties for configuring and managing camera models.

```APIDOC
## Camera Class

### create_from_model_id
Creates a camera instance using a model ID.
- **camera_id** (SupportsInt)
- **model** (pycolmap.CameraModelId)
- **focal_length** (SupportsFloat)
- **width** (SupportsInt)
- **height** (SupportsInt)

### create_from_model_name
Creates a camera instance using a model name string.
- **camera_id** (SupportsInt)
- **model_name** (str)
- **focal_length** (SupportsFloat)
- **width** (SupportsInt)
- **height** (SupportsInt)

### Properties
- **camera_id** (int)
- **model** (CameraModelId)
- **width** (int)
- **height** (int)
- **params** (ndarray)

### Methods
- **calibration_matrix()** -> Returns [3, 3] numpy array.
- **params_to_string()** -> Returns comma-separated string of parameters.
- **set_params_from_string(params: str)** -> Sets parameters from string, returns bool.
```