Add nms_rotated ort op (open-mmlab#312)

* fix pose demo and windows build (open-mmlab#307)

* init

* Update nms_rotated.cpp

* add postprocessing_masks gpu version (open-mmlab#276)

* add postprocessing_masks gpu version

* default device cpu

* pre-commit fix

Co-authored-by: hadoop-basecv <[email protected]>

* fixed a bug causes text-recognizer to fail when (non-NULL) empty bboxes list is passed (open-mmlab#310)

* [Fix] include missing <type_traits> for formatter.h (open-mmlab#313)

* fix formatter

* relax GCC version requirement

* fix

* fix lint

* fix lint

* [Fix] MMEditing cannot save results when testing (open-mmlab#336)

* fix show

* lint

* remove redundant codes

* resolve comment

* type hint

* docs(build): fix typo (open-mmlab#352)

* docs(build): add missing build option

* docs(build): add onnx install

* style(doc): trim whitespace

* docs(build): revert install onnx

* docs(build): add ncnn LD_LIBRARY_PATH

* docs(build): fix path error

* fix openvino export tmp model, add binary flag (open-mmlab#353)

* init circleci (open-mmlab#348)

* fix wrong input mat type (open-mmlab#362)

* fix wrong input mat type

* fix lint

* fix(docs): remove redundant doc tree (open-mmlab#360)

* fix missing ncnn_DIR & InferenceEngine_DIR (open-mmlab#364)

* update doc

Co-authored-by: Chen Xin <[email protected]>
Co-authored-by: Shengxi Li <[email protected]>
Co-authored-by: hadoop-basecv <[email protected]>
Co-authored-by: lzhangzz <[email protected]>
Co-authored-by: Yifan Zhou <[email protected]>
Co-authored-by: tpoisonooo <[email protected]>
Co-authored-by: lvhan028 <[email protected]>

.circleci/config.yml

@@ -0,0 +1,39 @@
+# Use the latest 2.1 version of CircleCI pipeline process engine.
+# See: https://circleci.com/docs/2.0/configuration-reference
+version: 2.1
+
+# Define a job to be invoked later in a workflow.
+# See: https://circleci.com/docs/2.0/configuration-reference/#jobs
+jobs:
+  lint:
+    # Specify the execution environment. You can specify an image from Dockerhub or use one of our Convenience Images from CircleCI's Developer Hub.
+    # See: https://circleci.com/docs/2.0/configuration-reference/#docker-machine-macos-windows-executor
+    docker:
+      - image: cimg/python:3.7.4
+    # Add steps to the job
+    # See: https://circleci.com/docs/2.0/configuration-reference/#steps
+    steps:
+      - checkout
+      - run:
+          name: Install pre-commit hook
+          command: |
+            sudo apt-add-repository ppa:brightbox/ruby-ng -y
+            sudo apt-get update
+            sudo apt-get install -y ruby2.7
+            pip install pre-commit
+            pre-commit install
+      - run:
+          name: Linting
+          command: pre-commit run --all-files
+      - run:
+          name: Check docstring coverage
+          command: |
+            pip install interrogate
+            interrogate -v --ignore-init-method --ignore-module --ignore-nested-functions --ignore-regex "__repr__" --fail-under 80 mmdeploy
+
+# Invoke jobs via workflows
+# See: https://circleci.com/docs/2.0/configuration-reference/#workflows
+workflows:
+  pr_stage_test:
+    jobs:
+      - lint

csrc/apis/c/classifier.cpp

@@ -86,8 +86,8 @@ int mmdeploy_classifier_apply(mm_handle_t handle, const mm_mat_t* mats, int mat_
 
     Value input{Value::kArray};
     for (int i = 0; i < mat_count; ++i) {
-      mmdeploy::Mat _mat{mats[i].height,       mats[i].width, PixelFormat(mats[i].format),
-                         DataType(mats->type), mats[i].data,  Device{"cpu"}};
+      mmdeploy::Mat _mat{mats[i].height,         mats[i].width, PixelFormat(mats[i].format),
+                         DataType(mats[i].type), mats[i].data,  Device{"cpu"}};
       input.front().push_back({{"ori_img", _mat}});
     }
 

csrc/apis/c/detector.cpp

@@ -85,8 +85,8 @@ int mmdeploy_detector_apply(mm_handle_t handle, const mm_mat_t* mats, int mat_co
 
     Value input{Value::kArray};
     for (int i = 0; i < mat_count; ++i) {
-      mmdeploy::Mat _mat{mats[i].height,       mats[i].width, PixelFormat(mats[i].format),
-                         DataType(mats->type), mats[i].data,  Device{"cpu"}};
+      mmdeploy::Mat _mat{mats[i].height,         mats[i].width, PixelFormat(mats[i].format),
+                         DataType(mats[i].type), mats[i].data,  Device{"cpu"}};
       input.front().push_back({{"ori_img", _mat}});
     }
 

csrc/apis/c/pose_detector.cpp

@@ -104,8 +104,8 @@ int mmdeploy_pose_detector_apply_bbox(mm_handle_t handle, const mm_mat_t* mats,
     Value input{Value::kArray};
     auto result_count = 0;
     for (int i = 0; i < mat_count; ++i) {
-      mmdeploy::Mat _mat{mats[i].height,       mats[i].width, PixelFormat(mats[i].format),
-                         DataType(mats->type), mats[i].data,  Device{"cpu"}};
+      mmdeploy::Mat _mat{mats[i].height,         mats[i].width, PixelFormat(mats[i].format),
+                         DataType(mats[i].type), mats[i].data,  Device{"cpu"}};
 
       Value img_with_boxes;
       if (bboxes && bbox_count) {

csrc/apis/c/segmentor.cpp

@@ -84,8 +84,8 @@ int mmdeploy_segmentor_apply(mm_handle_t handle, const mm_mat_t* mats, int mat_c
 
     Value input{Value::kArray};
     for (int i = 0; i < mat_count; ++i) {
-      mmdeploy::Mat _mat{mats[i].height,       mats[i].width, PixelFormat(mats[i].format),
-                         DataType(mats->type), mats[i].data,  Device{"cpu"}};
+      mmdeploy::Mat _mat{mats[i].height,         mats[i].width, PixelFormat(mats[i].format),
+                         DataType(mats[i].type), mats[i].data,  Device{"cpu"}};
       input.front().push_back({{"ori_img", _mat}});
     }
 

csrc/apis/c/text_recognizer.cpp

@@ -108,15 +108,21 @@ int mmdeploy_text_recognizer_apply_bbox(mm_handle_t handle, const mm_mat_t *imag
 
   try {
     auto recognizer = static_cast<Handle *>(handle);
-    Value input{Value::kArray, Value::kArray};
+    Value::Array input_images;
+    Value::Array input_bboxes;
     auto _bboxes = bboxes;
     auto result_count = 0;
-    for (int i = 0; i < image_count; ++i) {
-      mmdeploy::Mat _mat{images[i].height,       images[i].width, PixelFormat(images[i].format),
-                         DataType(images->type), images[i].data,  Device{"cpu"}};
-      input[0].push_back({{"ori_img", _mat}});
 
+    // mapping from image index to result index, -1 represents invalid image with no bboxes
+    // supplied.
+    std::vector<int> result_index(image_count, -1);
+
+    for (int i = 0; i < image_count; ++i) {
       if (bboxes && bbox_count) {
+        if (bbox_count[i] == 0) {
+          // skip images with no bounding boxes (push nothing)
+          continue;
+        }
         Value boxes(Value::kArray);
         for (int j = 0; j < bbox_count[i]; ++j) {
           Value box;
@@ -128,17 +134,26 @@ int mmdeploy_text_recognizer_apply_bbox(mm_handle_t handle, const mm_mat_t *imag
         }
         _bboxes += bbox_count[i];
         result_count += bbox_count[i];
-        input[1].push_back({{"boxes", boxes}});
+        input_bboxes.push_back({{"boxes", boxes}});
       } else {
-        input[1].push_back(Value::kNull);
+        // bboxes or bbox_count not supplied, use whole image
         result_count += 1;
+        input_bboxes.push_back(Value::kNull);
       }
+
+      result_index[i] = static_cast<int>(input_images.size());
+      mmdeploy::Mat _mat{images[i].height,         images[i].width, PixelFormat(images[i].format),
+                         DataType(images[i].type), images[i].data,  Device{"cpu"}};
+      input_images.push_back({{"ori_img", _mat}});
     }
 
-    auto output = recognizer->Run(std::move(input)).value().front();
+    std::vector<std::vector<mmocr::TextRecognizerOutput>> recognizer_outputs;
 
-    auto recognizer_outputs =
-        from_value<std::vector<std::vector<mmocr::TextRecognizerOutput>>>(output);
+    if (!input_images.empty()) {
+      Value input{std::move(input_images), std::move(input_bboxes)};
+      auto output = recognizer->Run(std::move(input)).value().front();
+      from_value(output, recognizer_outputs);
+    }
 
     std::vector<int> counts;
     if (bboxes && bbox_count) {
@@ -157,21 +172,23 @@ int mmdeploy_text_recognizer_apply_bbox(mm_handle_t handle, const mm_mat_t *imag
         new mm_text_recognize_t[result_count]{}, deleter);
 
     for (int i = 0; i < image_count; ++i) {
-      auto &recog_output = recognizer_outputs[i];
-      for (int j = 0; j < recog_output.size(); ++j) {
-        auto &res = _results[offsets[i] + j];
+      if (result_index[i] >= 0) {
+        auto &recog_output = recognizer_outputs[result_index[i]];
+        for (int j = 0; j < recog_output.size(); ++j) {
+          auto &res = _results[offsets[i] + j];
 
-        auto &box_result = recog_output[j];
+          auto &box_result = recog_output[j];
 
-        auto &score = box_result.score;
-        res.length = static_cast<int>(score.size());
+          auto &score = box_result.score;
+          res.length = static_cast<int>(score.size());
 
-        res.score = new float[score.size()];
-        std::copy_n(score.data(), score.size(), res.score);
+          res.score = new float[score.size()];
+          std::copy_n(score.data(), score.size(), res.score);
 
-        auto text = box_result.text;
-        res.text = new char[text.length() + 1];
-        std::copy_n(text.data(), text.length() + 1, res.text);
+          auto text = box_result.text;
+          res.text = new char[text.length() + 1];
+          std::copy_n(text.data(), text.length() + 1, res.text);
+        }
       }
     }
     *results = _results.release();