imcui.api.core

Core API module for image matching operations

imcui.api.core

Core API module for programmatic image matching operations.

ImageMatchingAPI

Main API class for image matching operations.

Import

from imcui.api import ImageMatchingAPI

Constructor

ImageMatchingAPI(conf, device="cuda")

Parameters:

Parameter	Type	Required	Default	Description
conf	dict	Yes	-	Matcher configuration from get_matcher_zoo()
device	str	No	”cuda”	Device: “cuda”, “cpu”, “mps”

Usage

from imcui.api import ImageMatchingAPI
from imcui.ui import get_matcher_zoo

# Get matcher configuration
matchers = get_matcher_zoo()

# Initialize API
api = ImageMatchingAPI(conf=matchers["superpoint-lightglue"], device="cuda")

# Match two images (RGB numpy arrays)
result = api(image0, image1)

Return Value

Dictionary containing:

{
    "keypoints0": np.ndarray,  # Shape: (N, 2)
    "keypoints1": np.ndarray,  # Shape: (M, 2)
    "matches": np.ndarray,     # Shape: (K, 2)
    "scores": np.ndarray,      # Shape: (K,)
    "H": np.ndarray            # Optional: Homography matrix (3x3)
}

Source Code: imcui/api/core.py