matcher.cuda_outlier_detector — Finding outliers in both the keypoints and the matches

The matcher.cpu_outlier_detector module

New in version 0.1.0.

autocnet.matcher.cuda_outlier_detector.distance_ratio(edge, matches, ratio=0.8)[source]

Compute and return a mask for a matches dataframe using Lowe’s ratio test. If keypoints have a single Lowe (2004) [Lowe2004]

Parameters
  • ratio (float) – the ratio between the first and second-best match distances for each keypoint to use as a bound for marking the first keypoint as “good”. Default: 0.8

  • single (bool) – If True, points with only a single entry are included (True) in the result mask, else False.

Returns

mask – A Pandas DataFrame mask for the matches with those failing the ratio test set to False.

Return type

pd.dataframe