Fusion

Custom functions can be passed to the fusion.fuse function. multiview-stitcher provides the custom fusion and weights functions with pre-transformed chunks corresponding to the transform_key passed to fusion.fuse. Only those views will be passed which contribute to the given output chunks. The elements of the input lists correspond to the different input views / tiles and conserve the order of the input views / tiles passed to fusion.fuse.

def fuse(
    images: list,
    ...
    transform_key: str = None,
    fusion_func: Callable = weighted_average_fusion,
    fusion_func_kwargs: dict = None,
    weights_func: Callable = None, # by default no additional fusion weights are used
    weights_func_kwargs: dict = None,
    ...

Custom fusion methods

Custom function for fusion of pre-transformed view chunks. This could implement e.g. a maximum intensity projection, a weighted average, or a multi-view deconvolution.

def custom_fusion_function(
    transformed_views: List[Array-like], # list of pre-transformed view chunks
    blending_weights: List[Array-like], # optional functional argument
    fusion_weights: List[Array-like], # optional functional argument
    **kwargs, # `fusion_func_kwargs` passed to `fusion.fuse`
) -> Array-like:

    # Fusion code here

    return fused_array

If the optional funtion arguments are not part of the function signature, the arguments will be ignored.

Example implementation: multiview_stitcher.fusion.weighted_average_fusion.

Custom weight functions

Custom function for calculating additional fusion weights passed to the fusion function as fusion_weights. This could implement e.g. a content-based weighting scheme. The function should return a list of weights, one for each view chunk.

def custom_weight_function(
    transformed_views : List[Array-like],
    blending_weights : List[Array-like], # optional functional argument
    params: List[Array-like], # optional functional argument
    **kwargs, # `weights_func_kwargs` passed to `fusion.fuse`
) - > List[Array-like]:

    # Weight calculation code here

    return weights

Example implementation: multiview_stitcher.weights.content_based.

As with custom fusion functions, optional arguments are only passed when they are declared in the function signature. In particular, blending_weights is only provided when the weight function accepts it.

CuPy support

If fuse(..., backend="cupy") is passed, the fusion and weight functions receive CuPy arrays as input and should return a CuPy array as output. The built-in fusion and weight functions support CuPy arrays (in large part by using the NumPy Array API which transparently dispatches functions to their numpy or cupy implementations based on the input array type).

See GPU support for further details and examples.