Tiling and TilesPatcher

.github/CODEOWNERS

@@ -0,0 +1 @@
+* @tersekmatija @jkbmrz @kkeroo @klemen1999

.github/labeler.yaml

@@ -0,0 +1,33 @@
+DevOps:
+  - changed-files:
+    - any-glob-to-any-file: ".github/*"
+documentation:
+  - changed-files:
+    - any-glob-to-any-file: "docs/*"
+examples:
+  - changed-files:
+    - any-glob-to-any-file: "examples/*"
+messages:
+  - changed-files:
+    - any-glob-to-any-file: "depthai_nodes/ml/messages/*"
+parsers:
+  - changed-files:
+    - any-glob-to-any-file: "depthai_nodes/ml/parsers/*"
+tests:
+  - changed-files:
+    - any-glob-to-any-file: "tests/*"
+enhancement:
+  - head-branch:
+    - 'feature/*'
+    - 'feat/*'
+    - 'enhancement/*'
+fix:
+  - head-branch:
+    - 'fix/*'
+    - 'bug/*'
+    - 'hotfix/*'
+    - 'issue/*'
+    - 'bugfix/*'
+    - 'patch/*'
+release:
+  - base-branch: 'main'

.github/workflows/ci.yaml

@@ -6,8 +6,9 @@ on:
     paths:
       - 'depthai_nodes/**'
       - 'tests/**'
-      - .github/workflows/ci.yaml
       - 'examples/**'
+      - 'docs/**'
+      - .github/workflows/ci.yaml
 
 permissions:
   pull-requests: write
@@ -21,6 +22,19 @@ jobs:
       - name: Auto-assign
         uses: toshimaru/auto-author-assign@v2.1.1
 
+  labeler:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.head_ref }}
+
+      - name: Labeler
+        uses: actions/labeler@v5
+        with:
+          configuration-path: .github/labeler.yaml
+
   pre-commit:
     runs-on: ubuntu-latest
     steps:
@@ -109,8 +123,8 @@ jobs:
       run: |
         git config --global user.name 'GitHub Actions'
         git config --global user.email 'actions@github.com'
-        git diff --quiet media/coverage_badge.svg || {
-          git add media/coverage_badge.svg
+        git add media/coverage_badge.svg
+        git diff --quiet --cached media/coverage_badge.svg || {
           git commit -m "[Automated] Updated coverage badge"
         }
 

.github/workflows/python-publish.yaml

@@ -0,0 +1,38 @@
+name: Upload Python Package
+
+on:
+  workflow_dispatch:
+  release:
+    types: [published]
+
+permissions:
+  contents: read
+
+jobs:
+  deploy:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v3
+
+    - name: Set up Python
+      uses: actions/setup-python@v3
+      with:
+        python-version: '3.8'
+        cache: pip
+
+    - name: Install dependencies
+      run: pip install build pydoctor
+
+    - name: Generate docs
+      run: |
+        curl -L "https://raw.githubusercontent.com/luxonis/python-api-analyzer-to-json/main/gen-docs.py" -o "gen-docs.py"
+        python gen-docs.py depthai_nodes
+
+    - name: Build package
+      run: python -m build
+
+    - name: Publish package
+      uses: pypa/gh-action-pypi-publish@27b31702a0e7fc50959f5ad993c78deac1bdfc29
+      with:
+        user: __token__
+        password: ${{ secrets.PYPI_API_TOKEN }}

.pre-commit-config.yaml

@@ -16,12 +16,6 @@ repos:
         additional_dependencies: [tomli]
         args: [--in-place, --black, --style=epytext]
 
-  - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
-    hooks:
-    - id: no-commit-to-branch
-      args: ['--branch', 'main', '--branch', 'dev']
-
   - repo: https://github.com/executablebooks/mdformat
     rev: 0.7.10
     hooks:

CONTRIBUTING.md

@@ -12,7 +12,7 @@ It outlines our workflow and standards for contributing to this project.
 
 ## Developing parser
 
-Parser should be developed so that it is consistent with other parsers. Check out other parsers to see the required structure. Additionally, pay attention to the naming of the parser's attributes. Check out [NN Archive Parameters](docs/nn_archive_parameters.md).
+Parser should be developed so that it is consistent with other parsers. Check out other parsers to see the required structure. Additionally, pay attention to the naming of the parser's attributes. Check out [Developer guide](docs/developer_guide.md).
 
 ## Pre-commit Hooks
 

README.md

@@ -2,6 +2,9 @@
 
 [![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
 
+![CI](https://github.com/luxonis/depthai-nodes/actions/workflows/ci.yaml/badge.svg)
+![Coverage](https://github.com/luxonis/depthai-nodes/blob/dev/media/coverage_badge.svg)
+
 [![Ruff](https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json)](https://github.com/astral-sh/ruff)
 [![Docformatter](https://img.shields.io/badge/%20formatter-docformatter-fedcba.svg)](https://github.com/PyCQA/docformatter)
 [![Black](https://img.shields.io/badge/code%20style-black-000000.svg)](https://github.com/psf/black)
@@ -33,18 +36,22 @@ To install the package, run:
 pip install depthai-nodes
 ```
 
-Before the official release on PyPI you can install the package from the GitHub repository:
+### Manual installation
+
+If you want to manually install the package from GitHub repositoory you can run:
 
 ```bash
 git clone git@github.com:luxonis/depthai-nodes.git
 ```
 
-and then install the requirements:
+and then inside the directory run:
 
 ```bash
-pip install -r requirements.txt
+pip install .
 ```
 
+Note: You'll still need to manually install `depthai v3`.
+
 ## Contributing
 
 If you want to contribute to this project, read the instructions in [CONTRIBUTING.md](./CONTRIBUTING.md)

depthai_nodes/__init__.py

@@ -1 +1,3 @@
 from .ml.parsers import *
+
+__version__ = "0.0.1"

depthai_nodes/ml/messages/creators/depth.py

@@ -7,7 +7,9 @@
 
 
 def create_depth_message(
-    depth_map: np.ndarray, depth_type: Literal["relative", "metric"]
+    depth_map: np.ndarray,
+    depth_type: Literal["relative", "metric"],
+    depth_limit: float = 0.0,
 ) -> dai.ImgFrame:
     """Create a DepthAI message for a depth map.
 
@@ -17,23 +19,28 @@ def create_depth_message(
     @param depth_type: A string indicating the type of depth map. It can either be
         'relative' or 'metric'.
     @type depth_type: Literal['relative', 'metric']
+    @param depth_limit: The maximum depth value (in meters) to be used in the depth map.
+        The default value is 0, which means no limit.
+    @type depth_limit: float
     @return: An ImgFrame object containing the depth information.
     @rtype: dai.ImgFrame
     @raise ValueError: If the depth map is not a NumPy array.
     @raise ValueError: If the depth map is not 2D or 3D.
     @raise ValueError: If the depth map shape is not NHW or HWN.
     @raise ValueError: If the depth type is not 'relative' or 'metric'.
-    @raise NotImplementedError: If the depth type is 'metric'.
+    @raise ValueError: If the depth limit is not 0 and the depth type is 'relative'.
+    @raise ValueError: If the depth limit is 0 and the depth type is 'metric'.
+    @raise ValueError: If the depth limit is negative.
     """
 
     if not isinstance(depth_map, np.ndarray):
         raise ValueError(f"Expected numpy array, got {type(depth_map)}.")
 
     if len(depth_map.shape) == 3:
         if depth_map.shape[0] == 1:
-            depth_map = depth_map[0, :, :]  # CHW to HW
+            depth_map = depth_map[0, :, :]  # NHW to HW
         elif depth_map.shape[2] == 1:
-            depth_map = depth_map[:, :, 0]  # HWC to HW
+            depth_map = depth_map[:, :, 0]  # HWN to HW
         else:
             raise ValueError(
                 f"Unexpected image shape. Expected NHW or HWN, got {depth_map.shape}."
@@ -42,27 +49,41 @@ def create_depth_message(
     if len(depth_map.shape) != 2:
         raise ValueError(f"Expected 2D or 3D input, got {len(depth_map.shape)}D input.")
 
-    if depth_type == "relative":
-        data_type = dai.ImgFrame.Type.RAW16
+    if not (depth_type == "relative" or depth_type == "metric"):
+        raise ValueError(
+            f"Invalid depth type: {depth_type}. Only 'relative' and 'metric' are supported."
+        )
 
-        # normalize depth map to the range [0, 65535]
-        min_val = depth_map.min()
-        max_val = depth_map.max()
-        if min_val == max_val:  # avoid division by zero
-            depth_map = np.zeros_like(depth_map)
-        else:
-            depth_map = (depth_map - min_val) / (max_val - min_val) * UINT16_MAX_VALUE
-        depth_map = depth_map.astype(np.uint16)
+    if depth_type == "relative" and depth_limit != 0:
+        raise ValueError(
+            f"Invalid depth limit: {depth_limit}. For relative depth, depth limit must be equal to 0."
+        )
 
-    elif depth_type == "metric":
-        raise NotImplementedError(
-            "The message for 'metric' depth type is not yet implemented."
+    if depth_type == "metric" and depth_limit == 0:
+        raise ValueError(
+            f"Invalid depth limit: {depth_limit}. For metric depth, depth limit must be bigger than 0."
         )
-    else:
+
+    if depth_limit < 0:
         raise ValueError(
-            f"Invalid depth type: {depth_type}. Only 'relative' and 'metric' are supported."
+            f"Invalid depth limit: {depth_limit}. Depth limit must be bigger than 0."
         )
 
+    data_type = dai.ImgFrame.Type.RAW16
+
+    min_val = depth_map.min() if depth_type == "relative" else 0
+    max_val = depth_map.max() if depth_type == "relative" else depth_limit
+
+    # clip values bigger than max_val
+    depth_map = np.clip(depth_map, a_min=None, a_max=max_val)
+
+    # normalize depth map to UINT16 range [0, UINT16_MAX_VALUE]
+    if min_val == max_val:  # avoid division by zero
+        depth_map = np.zeros_like(depth_map)
+    else:
+        depth_map = (depth_map - min_val) / (max_val - min_val) * UINT16_MAX_VALUE
+    depth_map = depth_map.astype(np.uint16)
+
     imgFrame = dai.ImgFrame()
     imgFrame.setFrame(depth_map)
     imgFrame.setWidth(depth_map.shape[1])

depthai_nodes/ml/parsers/mediapipe_palm_detection.py

@@ -23,6 +23,8 @@ class MPPalmDetectionParser(dai.node.ThreadedHostNode):
         Non-maximum suppression threshold.
     max_det : int
         Maximum number of detections to keep.
+    scale : int
+        Scale of the input image.
 
     Output Message/s
     -------
@@ -36,7 +38,7 @@ class MPPalmDetectionParser(dai.node.ThreadedHostNode):
     https://ai.google.dev/edge/mediapipe/solutions/vision/hand_landmarker
     """
 
-    def __init__(self, conf_threshold=0.5, iou_threshold=0.5, max_det=100):
+    def __init__(self, conf_threshold=0.5, iou_threshold=0.5, max_det=100, scale=192):
         """Initializes the MPPalmDetectionParser node.
 
         @param conf_threshold: Confidence score threshold for detected hands.
@@ -53,6 +55,7 @@ def __init__(self, conf_threshold=0.5, iou_threshold=0.5, max_det=100):
         self.conf_threshold = conf_threshold
         self.iou_threshold = iou_threshold
         self.max_det = max_det
+        self.scale = scale
 
     def setConfidenceThreshold(self, threshold):
         """Sets the confidence score threshold for detected hands.
@@ -78,26 +81,48 @@ def setMaxDetections(self, max_det):
         """
         self.max_det = max_det
 
+    def setScale(self, scale):
+        """Sets the scale of the input image.
+
+        @param scale: Scale of the input image.
+        @type scale: int
+        """
+        self.scale = scale
+
     def run(self):
         while self.isRunning():
             try:
                 output: dai.NNData = self.input.get()
             except dai.MessageQueue.QueueException:
                 break  # Pipeline was stopped
 
-            bboxes = (
-                output.getTensor("Identity", dequantize=True)
-                .reshape(2016, 18)
-                .astype(np.float32)
-            )
-            scores = (
-                output.getTensor("Identity_1", dequantize=True)
-                .reshape(2016)
-                .astype(np.float32)
-            )
+            all_tensors = output.getAllLayerNames()
+
+            bboxes = None
+            scores = None
+
+            for tensor_name in all_tensors:
+                tensor = output.getTensor(tensor_name, dequantize=True).astype(
+                    np.float32
+                )
+                if bboxes is None:
+                    bboxes = tensor
+                    scores = tensor
+                else:
+                    bboxes = bboxes if tensor.shape[-1] < bboxes.shape[-1] else tensor
+                    scores = tensor if tensor.shape[-1] < scores.shape[-1] else scores
+
+            bboxes = bboxes.reshape(-1, 18)
+            scores = scores.reshape(-1)
+
+            if bboxes is None or scores is None:
+                raise ValueError("No valid output tensors found.")
 
             decoded_bboxes = generate_anchors_and_decode(
-                bboxes=bboxes, scores=scores, threshold=self.conf_threshold, scale=192
+                bboxes=bboxes,
+                scores=scores,
+                threshold=self.conf_threshold,
+                scale=self.scale,
             )
 
             bboxes = []
@@ -123,6 +148,8 @@ def run(self):
             bboxes = np.array(bboxes)[indices]
             scores = np.array(scores)[indices]
 
+            bboxes = bboxes.astype(np.float32) / self.scale
+
             detections_msg = create_detection_message(bboxes, scores, labels=None)
             detections_msg.setTimestamp(output.getTimestamp())
             self.out.send(detections_msg)