gt dir: (wider_face_val.mat, wider_easy_val.mat, wider_medium_val.mat, wider_hard_val.mat)

Get the minimal bounding box of a group of points. The coordinates are also converted to int numbers.

Args: box_scores (N, 5): boxes in corner-form and probabilities. iou_threshold: intersection over union threshold. top_k

Args: box_scores (N, 5): boxes in corner-form and probabilities. iou_threshold: intersection over union threshold. top_k

Check if name is already used for graph inputs/outputs or for nodes inputs/outputs

single image evaluation pred: Nx5 gt: Nx4 ignore:

Return intersection-over-union (Jaccard index) of boxes. Args: boxes0 (N, 4): ground truth boxes. boxes1 (N or 1, 4): predicted b

Return intersection-over-union (Jaccard index) of boxes. Args: boxes0 (N, 4): ground truth boxes. boxes1 (N or 1, 4): predicted b

Returns true if this subset of nodes is eligible for fusion.

Compute the jaccard overlap of two sets of boxes. The jaccard overlap is simply the intersection over union of two boxes. E.g.: A ∩ B

Compute the jaccard overlap of two sets of boxes. The jaccard overlap is simply the intersection over union of two boxes. E.g.: A ∩ B

learning rate poly decay Args: base_lr: the base learning rate iter: the current iter Returns: the current learn

Merge nodes

norm score pred {key: [[x1,y1,x2,y2,s]]}

Compute the jaccard overlap of two sets of boxes. The jaccard overlap is simply the intersection over union of two boxes. E.g.: A ∩ B

Compute the jaccard overlap of two sets of boxes. The jaccard overlap is simply the intersection over union of two boxes. E.g.: A ∩ B

Find out the correspondence between layer names and parameter names.

FOR TRAIN LABELS raw labels form: [x,y,w,h, x1,y1,x2,y2, x3,y3,w3,h3, occ_type, occ_degree, gender, race, orientation, x4,y4,w4,h

Check if box contains all the points

Soft NMS implementation. References: https://arxiv.org/abs/1704.04503 https://github.com/facebookresearch/Detectron/blob/master/d

Soft NMS implementation. References: https://arxiv.org/abs/1704.04503 https://github.com/facebookresearch/Detectron/blob/master/d

Args: image (Tensor): image tensor to be transformed Return: a tensor with channels swapped according to swap

Args: img: the output of cv.imread in RGB layout. boxes: boundding boxes in the form of (x1, y1, x2, y2).