[Model] Support PP-StructureV2-Layout model (#1867)

* [Model] init pp-structurev2-layout code * [Model] init pp-structurev2-layout code * [Model] init pp-structurev2-layout code * [Model] add structurev2_layout_preprocessor * [PP-StructureV2] add postprocessor and layout detector class * [PP-StructureV2] add postprocessor and layout detector class * [PP-StructureV2] add postprocessor and layout detector class * [PP-StructureV2] add postprocessor and layout detector class * [PP-StructureV2] add postprocessor and layout detector class * [pybind] add pp-structurev2-layout model pybind * [pybind] add pp-structurev2-layout model pybind * [Bug Fix] fixed code style * [examples] add pp-structurev2-layout c++ examples * [PP-StructureV2] add python example and docs * [benchmark] add pp-structurev2-layout benchmark support
2025-10-06 00:57:33 +08:00 · 2023-05-05 13:05:58 +08:00
parent 2c5fd91a7f
commit 6d0261e9e4
26 changed files with 1255 additions and 23 deletions
--- a/examples/vision/ocr/PP-OCR/cpu-gpu/python/infer_structurev2_layout.py
+++ b/examples/vision/ocr/PP-OCR/cpu-gpu/python/infer_structurev2_layout.py
@@ -0,0 +1,91 @@
+# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#    http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import fastdeploy as fd
+import cv2
+import os
+
+
+def parse_arguments():
+    import argparse
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--layout_model",
+        required=True,
+        help="Path of Layout detection model of PP-StructureV2.")
+    parser.add_argument(
+        "--image", type=str, required=True, help="Path of test image file.")
+    parser.add_argument(
+        "--device",
+        type=str,
+        default='cpu',
+        help="Type of inference device, support 'cpu' or 'gpu'.")
+    parser.add_argument(
+        "--device_id",
+        type=int,
+        default=0,
+        help="Define which GPU card used to run model.")
+
+    return parser.parse_args()
+
+
+def build_option(args):
+
+    layout_option = fd.RuntimeOption()
+
+    if args.device.lower() == "gpu":
+        layout_option.use_gpu(args.device_id)
+
+    return layout_option
+
+
+args = parse_arguments()
+
+layout_model_file = os.path.join(args.layout_model, "model.pdmodel")
+layout_params_file = os.path.join(args.layout_model, "model.pdiparams")
+
+# Set the runtime option
+layout_option = build_option(args)
+
+# Create the table_model
+layout_model = fd.vision.ocr.StructureV2Layout(
+    layout_model_file, layout_params_file, layout_option)
+
+layout_model.postprocessor.num_class = 5
+
+# Read the image
+im = cv2.imread(args.image)
+
+# Predict and return the results
+result = layout_model.predict(im)
+
+print(result)
+
+# Visualize the results
+labels = ["text", "title", "list", "table", "figure"]
+if layout_model.postprocessor.num_class == 10:
+    labels = [
+        "text", "title", "figure", "figure_caption", "table", "table_caption",
+        "header", "footer", "reference", "equation"
+    ]
+
+vis_im = fd.vision.vis_detection(
+    im,
+    result,
+    labels,
+    score_threshold=0.5,
+    font_color=[255, 0, 0],
+    font_thickness=2)
+cv2.imwrite("visualized_result.jpg", vis_im)
+print("Visualized result save in ./visualized_result.jpg")