Expose ROI Pooling into Python API.

python/paddle/fluid/layers/nn.py

@@ -79,6 +79,7 @@
     'lrn',
     'pad',
     'label_smooth',
+    'roi_pool',
 ]
 
 
@@ -3759,3 +3760,44 @@ def label_smooth(label,
         outputs={"Out": smooth_label},
         attrs={"epsilon": float(epsilon)})
     return smooth_label
+
+
+def roi_pool(input, rois, pooled_height=1, pooled_width=1, spatial_scale=1.0):
+    """
+    RoI pooling, its purpose is to perform max pooling on inputs of nonuniform sizes to obtain fixed-size feature maps (e.g. 7*7).
+    The operator has three steps: 
+        1.Dividing each region proposal into equal-sized sections with the pooled_width and pooled_height 
+        2.Finding the largest value in each section 
+        3.Copying these max values to the output buffer
+
+    Args:
+        input (Variable): The input for roi pooling.
+        rois (Variable): ROIs (Regions of Interest) to pool over. It should be a 2-D tensor of shape (num_rois, 5). It given as [[batch_id, x1, y1, x2, y2], ...], where batch_id is the id of the data, (x1, y1) is the top left coordinates, and (x2, y2) is the bottom right coordinates.
+        pooled_height (integer): The pooled output height. Default: 1
+        pooled_width (integer): The pooled output width. Default: 1
+        spatial_scale (float): Multiplicative spatial scale factor. To translate ROI coords from their input scale to the scale used when pooling. Default: 1.0
+
+    Returns:
+        pool_out (Variable): The output of ROIPoolOp is a 4-D tensor with shape (num_rois, channels, pooled_h, pooled_w).
+        argmaxes (Variable): Argmaxes corresponding to indices in input with shape (num_rois, channels, pooled_h, pooled_w). 
+
+    Examples:
+        .. code-block:: python
+             # assuming we have input x_feas, rois x_rois, pooled_height ph, pooled_width pw and spatial_scale scale.
+             pool_out, argmaxes = fluid.layers.roi_pool(input=x_feas, rois=x_rois, pooled_height=ph, pooled_width=pw, spatial_scale=scale)
+    """
+    helper = LayerHelper('roi_pool', **locals())
+    dtype = helper.input_dtype()
+    pool_out = helper.create_tmp_variable(dtype)
+    helper.append_op(
+        type="pool2d",
+        inputs={"X": input,
+                "ROIs": rois},
+        outputs={"Out": pool_out,
+                 "Argmax": argmaxes},
+        attrs={
+            "pooled_height": pooled_height,
+            "pooled_width": pooled_width,
+            "spatial_scale": spatial_scale
+        })
+    return pool_out, argmaxes

python/paddle/fluid/tests/unittests/test_layers.py

@@ -359,6 +359,17 @@ def test_topk(self):
             self.assertIsNotNone(indices)
         print(str(program))
 
+    def test_roi_pool(self):
+        print("test_roi_pool")
+        program = Program()
+        with program_guard(program):
+            x = layers.data(name="x", shape=[10, 256, 30, 30], dtype="float32")
+            rois = layers.data(name="rois", shape=[20, 5], dtype="float32")
+            output, argmaxes = layers.roi_pool(x, rois, 7, 7, 0.6)
+            self.assertIsNotNone(output)
+            self.assertIsNotNone(argmaxes)
+        print(str(program))
+
 
 if __name__ == '__main__':
     unittest.main()