utils.box#
General methods for working with boxes
For more details see https://d2l.ai/chapter_computer-vision/anchor.html.
Functions
Convert from (center, width, height) to (upper-left, lower-right) |
|
Convert from (upper-left, lower-right) to (center, width, height) |
|
Compute pairwise IoU across two lists of anchor or bounding boxes |
|
Predict bounding boxes using non-maximum suppression |
|
Sort confidence scores of predicted bounding boxes |
|
Transform for anchor box offsets |
|
Predict bounding boxes based on anchor boxes with predicted offsets |