Initial commit for AXERA AI accelerators

.github/workflows/ci.yml

@@ -225,3 +225,29 @@ jobs:
           sources: |
             ghcr.io/${{ steps.lowercaseRepo.outputs.lowercase }}:${{ env.SHORT_SHA }}-amd64
             ghcr.io/${{ steps.lowercaseRepo.outputs.lowercase }}:${{ env.SHORT_SHA }}-rpi
+  axera_build:
+    runs-on: ubuntu-22.04
+    name: AXERA Build
+    needs:
+      - amd64_build
+      - arm64_build
+    steps:
+      - name: Check out code
+        uses: actions/checkout@v5
+        with:
+          persist-credentials: false
+      - name: Set up QEMU and Buildx
+        id: setup
+        uses: ./.github/actions/setup
+        with:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      - name: Build and push Axera build
+        uses: docker/bake-action@v6
+        with:
+          source: .
+          push: true
+          targets: axcl
+          files: docker/axcl/axcl.hcl
+          set: |
+            axcl.tags=${{ steps.setup.outputs.image-name }}-axcl
+            *.cache-from=type=gha

docker/axcl/Dockerfile

@@ -0,0 +1,59 @@
+# syntax=docker/dockerfile:1.6
+
+# https://askubuntu.com/questions/972516/debian-frontend-environment-variable
+ARG DEBIAN_FRONTEND=noninteractive
+
+# Globally set pip break-system-packages option to avoid having to specify it every time
+ARG PIP_BREAK_SYSTEM_PACKAGES=1
+
+
+FROM frigate AS frigate-axcl
+ARG TARGETARCH
+ARG PIP_BREAK_SYSTEM_PACKAGES
+
+# Install axmodels
+RUN mkdir -p /axmodels \
+    && wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/yolov5s_320.axmodel -O /axmodels/yolov5s_320.axmodel
+
+# Install axpyengine
+RUN wget https://github.com/AXERA-TECH/pyaxengine/releases/download/0.1.3.rc1/axengine-0.1.3-py3-none-any.whl -O /axengine-0.1.3-py3-none-any.whl
+RUN pip3 install -i https://mirrors.aliyun.com/pypi/simple/ /axengine-0.1.3-py3-none-any.whl \
+    && rm /axengine-0.1.3-py3-none-any.whl
+
+# Install axcl
+RUN if [ "$TARGETARCH" = "amd64" ]; then \
+        echo "Installing x86_64 version of axcl"; \
+        wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/axcl_host_x86_64_V3.6.5_20250908154509_NO4973.deb -O /axcl.deb; \
+    else \
+        echo "Installing aarch64 version of axcl"; \
+        wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/axcl_host_aarch64_V3.6.5_20250908154509_NO4973.deb -O /axcl.deb; \
+    fi
+
+RUN mkdir /unpack_axcl && \
+    dpkg-deb -x /axcl.deb /unpack_axcl && \
+    cp -R /unpack_axcl/usr/bin/axcl /usr/bin/ && \
+    cp -R /unpack_axcl/usr/lib/axcl /usr/lib/ && \
+    rm -rf /unpack_axcl /axcl.deb
+
+
+# Install axcl ffmpeg
+RUN mkdir -p /usr/lib/ffmpeg/axcl
+
+RUN if [ "$TARGETARCH" = "amd64" ]; then \
+        wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/ffmpeg-x64 -O /usr/lib/ffmpeg/axcl/ffmpeg && \
+        wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/ffprobe-x64 -O /usr/lib/ffmpeg/axcl/ffprobe; \
+    else \
+        wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/ffmpeg-aarch64 -O /usr/lib/ffmpeg/axcl/ffmpeg && \
+        wget https://github.com/ivanshi1108/assets/releases/download/v0.16.2/ffprobe-aarch64 -O /usr/lib/ffmpeg/axcl/ffprobe; \
+    fi
+
+RUN chmod +x /usr/lib/ffmpeg/axcl/ffmpeg /usr/lib/ffmpeg/axcl/ffprobe
+
+# Set ldconfig path
+RUN echo "/usr/lib/axcl" > /etc/ld.so.conf.d/ax.conf
+
+# Set env
+ENV PATH="$PATH:/usr/bin/axcl"
+ENV LD_LIBRARY_PATH="$LD_LIBRARY_PATH:/usr/lib/axcl"
+
+ENTRYPOINT ["sh", "-c", "ldconfig && exec /init"]

docker/axcl/axcl.hcl

@@ -0,0 +1,13 @@
+target frigate {
+  dockerfile = "docker/main/Dockerfile"
+  platforms = ["linux/amd64", "linux/arm64"]
+  target = "frigate"
+}
+
+target axcl {
+  dockerfile = "docker/axcl/Dockerfile"
+  contexts = {
+    frigate = "target:frigate",
+  }
+  platforms = ["linux/amd64", "linux/arm64"]
+}

docker/axcl/axcl.mk

@@ -0,0 +1,15 @@
+BOARDS += axcl
+
+local-axcl: version
+	docker buildx bake --file=docker/axcl/axcl.hcl axcl \
+		--set axcl.tags=frigate:latest-axcl \
+		--load
+
+build-axcl: version
+	docker buildx bake --file=docker/axcl/axcl.hcl axcl \
+		--set axcl.tags=$(IMAGE_REPO):${GITHUB_REF_NAME}-$(COMMIT_HASH)-axcl
+
+push-axcl: build-axcl
+	docker buildx bake --file=docker/axcl/axcl.hcl axcl \
+		--set axcl.tags=$(IMAGE_REPO):${GITHUB_REF_NAME}-$(COMMIT_HASH)-axcl \
+		--push

docker/axcl/user_installation.sh

@@ -0,0 +1,83 @@
+#!/bin/bash
+
+# Update package list and install dependencies
+sudo apt-get update
+sudo apt-get install -y build-essential cmake git wget pciutils kmod udev
+
+# Check if gcc-12 is needed
+current_gcc_version=$(gcc --version | head -n1 | awk '{print $NF}')
+gcc_major_version=$(echo $current_gcc_version | cut -d'.' -f1)
+
+if [[ $gcc_major_version -lt 12 ]]; then
+    echo "Current GCC version ($current_gcc_version) is lower than 12, installing gcc-12..."
+    sudo apt-get install -y gcc-12
+    sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-12 12
+    echo "GCC-12 installed and set as default"
+else
+    echo "Current GCC version ($current_gcc_version) is sufficient, skipping GCC installation"
+fi
+
+# Determine architecture
+arch=$(uname -m)
+download_url=""
+
+if [[ $arch == "x86_64" ]]; then
+    download_url="https://github.com/ivanshi1108/assets/releases/download/v0.16.2/axcl_host_x86_64_V3.6.5_20250908154509_NO4973.deb"
+    deb_file="axcl_host_x86_64_V3.6.5_20250908154509_NO4973.deb"
+elif [[ $arch == "aarch64" ]]; then
+    download_url="https://github.com/ivanshi1108/assets/releases/download/v0.16.2/axcl_host_aarch64_V3.6.5_20250908154509_NO4973.deb"
+    deb_file="axcl_host_aarch64_V3.6.5_20250908154509_NO4973.deb"
+else
+    echo "Unsupported architecture: $arch"
+    exit 1
+fi
+
+# Download AXCL driver
+echo "Downloading AXCL driver for $arch..."
+wget "$download_url" -O "$deb_file"
+
+if [ $? -ne 0 ]; then
+    echo "Failed to download AXCL driver"
+    exit 1
+fi
+
+# Install AXCL driver
+echo "Installing AXCL driver..."
+sudo dpkg -i "$deb_file"
+
+if [ $? -ne 0 ]; then
+    echo "Failed to install AXCL driver, attempting to fix dependencies..."
+    sudo apt-get install -f -y
+    sudo dpkg -i "$deb_file"
+
+    if [ $? -ne 0 ]; then
+        echo "AXCL driver installation failed"
+        exit 1
+    fi
+fi
+
+# Update environment
+echo "Updating environment..."
+source /etc/profile
+
+# Verify installation
+echo "Verifying AXCL installation..."
+if command -v axcl-smi &> /dev/null; then
+    echo "AXCL driver detected, checking AI accelerator status..."
+
+    axcl_output=$(axcl-smi 2>&1)
+    axcl_exit_code=$?
+
+    echo "$axcl_output"
+
+    if [ $axcl_exit_code -eq 0 ]; then
+        echo "AXCL driver installation completed successfully!"
+    else
+        echo "AXCL driver installed but no AI accelerator detected or communication failed."
+        echo "Please check if the AI accelerator is properly connected and powered on."
+        exit 1
+    fi
+else
+    echo "axcl-smi command not found. AXCL driver installation may have failed."
+    exit 1
+fi

docs/docs/configuration/object_detectors.md

@@ -47,6 +47,11 @@ Frigate supports multiple different detectors that work on different types of ha
 
 - [Synaptics](#synaptics): synap models can run on Synaptics devices(e.g astra machina) with included NPUs.
 
+**AXERA**
+
+- [AXEngine](#axera): axmodels can run on AXERA AI acceleration.
+
+
 **For Testing**
 
 - [CPU Detector (not recommended for actual use](#cpu-detector-not-recommended): Use a CPU to run tflite model, this is not recommended and in most cases OpenVINO can be used in CPU mode with better results.
@@ -1099,6 +1104,40 @@ model:  # required
   labelmap_path: /labelmap/coco-80.txt  # required
 ```
 
+## AXERA
+
+Hardware accelerated object detection is supported on the following SoCs:
+
+- AX650N
+- AX8850N
+
+This implementation uses the [AXera Pulsar2 Toolchain](https://huggingface.co/AXERA-TECH/Pulsar2).
+
+See the [installation docs](../frigate/installation.md#axera) for information on configuring the AXEngine hardware.
+
+### Configuration
+
+When configuring the AXEngine detector, you have to specify the model name.
+
+#### yolov5s
+
+A yolov5s model is provided in the container at /axmodels and is used by this detector type by default. 
+
+Use the model configuration shown below when using the axengine detector with the default axmodel:
+
+```yaml
+detectors:  # required
+  axengine:  # required
+    type: axengine  # required
+
+model:  # required
+  path: yolov5s_320  # required
+  width: 320  # required
+  height: 320  # required
+  tensor_format: bgr  # required
+  labelmap_path: /labelmap/coco-80.txt  # required
+```
+
 ## Rockchip platform
 
 Hardware accelerated object detection is supported on the following SoCs:

docs/docs/frigate/hardware.md

@@ -110,6 +110,14 @@ Frigate supports multiple different detectors that work on different types of ha
 | ssd mobilenet    | ~ 25 ms                         |
 | yolov5m          | ~ 118 ms                        |
 
+### AXERA
+
+- **AXEngine** Default model is **yolov5s_320**
+
+| Name             | AXERA AX650N/AX8850N Inference Time |
+| ---------------- | ----------------------------------- |
+| yolov5s_320      | ~ 1.676 ms                          |
+
 ### Hailo-8
 
 Frigate supports both the Hailo-8 and Hailo-8L AI Acceleration Modules on compatible hardware platforms—including the Raspberry Pi 5 with the PCIe hat from the AI kit. The Hailo detector integration in Frigate automatically identifies your hardware type and selects the appropriate default model when a custom model isn’t provided.

docs/docs/frigate/installation.md

@@ -287,6 +287,40 @@ or add these options to your `docker run` command:
 
 Next, you should configure [hardware object detection](/configuration/object_detectors#synaptics) and [hardware video processing](/configuration/hardware_acceleration_video#synaptics).
 
+### AXERA
+
+AXERA accelerators are available in an M.2 form factor, compatible with both Raspberry Pi and Orange Pi. This form factor has also been successfully tested on x86 platforms, making it a versatile choice for various computing environments.
+
+#### Installation
+
+Using AXERA accelerators requires the installation of the AXCL driver. We provide a convenient Linux script to complete this installation.
+
+Follow these steps for installation:
+
+1. Copy or download [this script](https://github.com/ivanshi1108/assets/releases/download/v0.16.2/user_installation.sh).
+2. Ensure it has execution permissions with `sudo chmod +x user_installation.sh`
+3. Run the script with `./user_installation.sh`
+
+#### Setup
+
+To set up Frigate, follow the default installation instructions, for example: `ghcr.io/blakeblackshear/frigate:stable`
+
+Next, grant Docker permissions to access your hardware by adding the following lines to your `docker-compose.yml` file:
+
+```yaml
+devices:
+  - /dev/axcl_host
+  - /dev/ax_mmb_dev
+  - /dev/msg_userdev
+```
+
+If you are using `docker run`, add this option to your command `--device /dev/axcl_host --device /dev/ax_mmb_dev --device /dev/msg_userdev`
+
+#### Configuration
+
+Finally, configure [hardware object detection](/configuration/object_detectors#axera) to complete the setup.
+
+
 ## Docker
 
 Running through Docker with Docker Compose is the recommended install method.