search

pythonlessons / TensorFlow-2.x-YOLOv3

YOLOv3 implementation in TensorFlow 2.3.1
https://pylessons.com/
MIT License
605 stars 327 forks source link

ciou_enclose_area #157

Open lihualilee opened 3 years ago

lihualilee commented 3 years ago

image left and up why not tf.minimum()？why mot computer minimum is compute enclose_area? left = tf.maximum(boxes1_coor[..., 0], boxes2_coor[..., 0]) up = tf.maximum(boxes1_coor[..., 1], boxes2_coor[..., 1])

pythonlessons commented 2 years ago

Unfortunately, this was not used in a code yet.

But you can see that we use: left - [..., 0], up - [..., 1] ...

    left = tf.maximum(boxes1_coor[..., 0], boxes2_coor[..., 0])
    up = tf.maximum(boxes1_coor[..., 1], boxes2_coor[..., 1])
    right = tf.maximum(boxes1_coor[..., 2], boxes2_coor[..., 2])
    down = tf.maximum(boxes1_coor[..., 3], boxes2_coor[..., 3])

this is why we use maximum