traccuracy.matchers._compute_overlap

Fast R-CNN via numba

adapted from Fast R-CNN Written by Sergey Karayev Licensed under The MIT License [see LICENSE for details] Copyright (c) 2015 Microsoft

Module Contents

Functions

graph_bbox_and_labels(→ tuple[numpy.ndarray | None, ...)

Get bounding boxes and labels for a list of nodes in a graph.

get_labels_with_overlap(→ list[tuple[int, int, float]])

Get all labels IDs in gt_frame and res_frame whose bounding boxes overlap,

compute_overlap(→ numpy.ndarray)

Args

compute_overlap_3D(→ numpy.ndarray)

Args

Attributes

signature

traccuracy.matchers._compute_overlap.graph_bbox_and_labels(graph: networkx.DiGraph, nodes: collections.abc.Iterable[collections.abc.Hashable], label_key: str | None = 'segmentation_id') → tuple[numpy.ndarray | None, numpy.ndarray | None][source]

Get bounding boxes and labels for a list of nodes in a graph. If a node is missing the ‘bbox’ or ‘segmentation_id’ attributes, it returns None for both bounding boxes and labels.

Parameters:

  • graph (nx.DiGraph) – The graph to get the bounding boxes and labels from.

  • nodes (list[Hashable]) – The nodes to get the bounding boxes and labels for.

  • label_key (str, optional) – The key to use for the labels. Defaults to ‘segmentation_id’.

Returns:

The bounding boxes and labels for the nodes.

Return type:

tuple[np.ndarray | None, np.ndarray | None]

traccuracy.matchers._compute_overlap.get_labels_with_overlap(gt_frame: numpy.ndarray, res_frame: numpy.ndarray, gt_boxes: numpy.ndarray | None = None, res_boxes: numpy.ndarray | None = None, gt_labels: numpy.ndarray | None = None, res_labels: numpy.ndarray | None = None, overlap: str = 'iou') → list[tuple[int, int, float]][source]

Get all labels IDs in gt_frame and res_frame whose bounding boxes overlap, and a metric of pixel overlap (either iou or iogt).

Parameters:

  • gt_frame (np.ndarray) – ground truth segmentation for a single frame

  • res_frame (np.ndarray) – result segmentation for a given frame

  • gt_boxes (np.ndarray) – ground truth bounding boxes for a single frame

  • res_boxes (np.ndarray) – result bounding boxes for a given frame

  • gt_labels (np.ndarray) – ground truth labels for a single frame

  • res_labels (np.ndarray) – result labels for a given frame

  • overlap (str, optional) – Choose between intersection-over-ground-truth (iogt) or intersection-over-union (iou). Defaults to iou.

Returns: list[tuple[int, int, float]] A list of tuples of overlapping labels and their

overlap values. Each tuple contains (gt_label, res_label, overlap_value).

traccuracy.matchers._compute_overlap.compute_overlap(boxes: numpy.ndarray, query_boxes: numpy.ndarray) → numpy.ndarray[source]
Args

boxes: (N, 4) ndarray of float query_boxes: (K, 4) ndarray of float

Returns

overlaps: (N, K) ndarray of overlap between boxes and query_boxes

traccuracy.matchers._compute_overlap.compute_overlap_3D(boxes: numpy.ndarray, query_boxes: numpy.ndarray) → numpy.ndarray[source]
Args

boxes: (N, 6) ndarray of float query_boxes: (K, 6) ndarray of float

Returns

overlaps: (N, K) ndarray of overlap between boxes and query_boxes

traccuracy.matchers._compute_overlap.signature