first commit

ssd/modeling/anchors/prior_box.py

@@ -0,0 +1,54 @@
+from itertools import product
+
+import torch
+from math import sqrt
+
+
+class PriorBox:
+    def __init__(self, cfg):
+        self.image_size = cfg.INPUT.IMAGE_SIZE #300
+        prior_config = cfg.MODEL.PRIORS
+        self.feature_maps = prior_config.FEATURE_MAPS
+        self.min_sizes = prior_config.MIN_SIZES  #[30, 60, 111, 162, 213, 264]
+        self.max_sizes = prior_config.MAX_SIZES  #[60, 111, 162, 213, 264, 315]
+        self.strides = prior_config.STRIDES #[8, 16, 32, 64, 100, 300]
+        self.aspect_ratios = prior_config.ASPECT_RATIOS #[[2], [2, 3], [2, 3], [2, 3], [2], [2]]
+        self.clip = prior_config.CLIP
+
+    def __call__(self):
+        """Generate SSD Prior Boxes.
+            It returns the center, height and width of the priors. The values are relative to the image size
+            Returns:
+                priors (num_priors, 4): The prior boxes represented as [[center_x, center_y, w, h]]. All the values
+                    are relative to the image size.
+        """
+        priors = []
+        for k, f in enumerate(self.feature_maps):
+            scale = self.image_size / self.strides[k]
+            for i, j in product(range(f), repeat=2):
+                # unit center x,y
+                cx = (j + 0.5) / scale
+                cy = (i + 0.5) / scale
+
+                # small sized square box
+                size = self.min_sizes[k]
+                h = w = size / self.image_size
+                priors.append([cx, cy, w, h])
+
+                # big sized square box
+                size = sqrt(self.min_sizes[k] * self.max_sizes[k])
+                h = w = size / self.image_size
+                priors.append([cx, cy, w, h])
+
+                # change h/w ratio of the small sized box
+                size = self.min_sizes[k]
+                h = w = size / self.image_size
+                for ratio in self.aspect_ratios[k]:
+                    ratio = sqrt(ratio)
+                    priors.append([cx, cy, w * ratio, h / ratio])
+                    priors.append([cx, cy, w / ratio, h * ratio])
+
+        priors = torch.tensor(priors)
+        if self.clip:
+            priors.clamp_(max=1, min=0)       #限制priors的范围在0-1之间
+        return priors