add pspnet describ

pdseg/models/modeling/pspnet.py

@@ -12,6 +12,7 @@ from models.backbone.resnet import ResNet as resnet_backbone
 from utils.config import cfg
 
 def get_logit_interp(input, num_classes, out_shape, name="logit"):
+    # 根据类别数决定最后一层卷积输出, 并插值回原始尺寸
     param_attr = fluid.ParamAttr(
         name=name + 'weights',
         regularizer=fluid.regularizer.L2DecayRegularizer(
@@ -19,13 +20,12 @@ def get_logit_interp(input, num_classes, out_shape, name="logit"):
         initializer=fluid.initializer.TruncatedNormal(loc=0.0, scale=0.01))
 
     with scope(name):
-        logit = conv(
-            input,
-            num_classes,
-            filter_size=1,
-            param_attr=param_attr,
-            bias_attr=True,
-            name=name+'.conv2d.output.1')
+        logit = conv(input,
+                    num_classes,
+                    filter_size=1,
+                    param_attr=param_attr,
+                    bias_attr=True,
+                    name=name+'_conv')
         logit_interp = fluid.layers.resize_bilinear(
                     logit, 
                     out_shape=out_shape,
@@ -34,53 +34,67 @@ def get_logit_interp(input, num_classes, out_shape, name="logit"):
 
 
 def psp_module(input, out_features):
+    # Pyramid Scene Parsing 金字塔池化模块
+    # 输入：backbone输出的特征
+    # 输出：对输入进行不同尺度pooling, 卷积操作后插值回原始尺寸，并concat
+    #       最后进行一个卷积及BN操作
+    
     cat_layers = []
     sizes = (1,2,3,6)
     for size in sizes:
         psp_name = "psp_conv" + str(size)
         with scope(psp_name):
             pool = fluid.layers.adaptive_pool2d(input, 
-                pool_size=[size, size], 
-                pool_type='avg', 
-                name=psp_name+'_adapool')
-            data = conv(pool, out_features, filter_size=1, bias_attr=True, 
-                    name= psp_name + '.conv2d.output.1')
+                    pool_size=[size, size], 
+                    pool_type='avg', 
+                    name=psp_name+'_adapool')
+            data = conv(pool, out_features, 
+                    filter_size=1,
+                    bias_attr=True, 
+                    name= psp_name + '_conv')
             data_bn = bn(data, act='relu')
             interp = fluid.layers.resize_bilinear(data_bn, 
-                out_shape=input.shape[2:], 
-                name=psp_name+'_interp') 
+                    out_shape=input.shape[2:], 
+                    name=psp_name+'_interp') 
         cat_layers.append(interp)
     cat_layers = [input] + cat_layers[::-1]
     cat = fluid.layers.concat(cat_layers, axis=1, name='psp_cat')
-    with scope("psp_conv_end"):
+
+    psp_end_name = "psp_conv_end"
+    with scope(psp_end_name):
         data = conv(cat, 
-                out_features, 
-                filter_size=3,
-                padding=1, 
-                bias_attr=True,
-                name='psp_conv_end.conv2d.output.1')
+                    out_features, 
+                    filter_size=3,
+                    padding=1, 
+                    bias_attr=True,
+                    name=psp_end_name)
         out = bn(data, act='relu')
 
     return out
 
 def resnet(input):
-    # PSPNET backbone: resnet, ĬÈresnet50
-    # end_points: resnetÖֹ²ã
-
+    # PSPNET backbone: resnet, 默认resnet50
+    # end_points: resnet终止层数
+    # dilation_dict: resnet block数及对应的膨胀卷积尺度
+    scale = cfg.MODEL.ICNET.DEPTH_MULTIPLIER
     scale = cfg.MODEL.PSPNET.DEPTH_MULTIPLIER
     layers = cfg.MODEL.PSPNET.LAYERS
     end_points = layers - 1
     dilation_dict = {2:2, 3:4}
     model = resnet_backbone(layers, scale, stem='pspnet')
-    data, _ = model.net(input, end_points=end_points, dilation_dict=dilation_dict)
+    data, _ = model.net(input, 
+                        end_points=end_points, 
+                        dilation_dict=dilation_dict)
 
     return data
 
 def pspnet(input, num_classes):
+    # Backbone: ResNet
     res = resnet(input)
+    # PSP模块
     psp = psp_module(res, 512)
-    #dropout = fluid.layers.dropout(psp, dropout_prob=0.1, name="dropout")
-    logit = get_logit_interp(psp, num_classes, input.shape[2:])
+    dropout = fluid.layers.dropout(psp, dropout_prob=0.1, name="dropout")
+    # 根据类别数决定最后一层卷积输出, 并插值回原始尺寸
+    logit = get_logit_interp(dropout, num_classes, input.shape[2:])
     return logit
 
-