CV / STDC1

STDC1

STDC1 (Short-Term Dense Concatenate) führt semantische Segmentierung mit 19 Cityscapes-Klassen über den HailoRT-Inferenzdienst aus.

12 Downloads

Größe

11 MB

Speicher

4GB+

Präzision

Hailo HEF / HailoRT

Wähle das Gerät, das du verwendest. Die Einrichtungsanleitung und Dokumentation werden entsprechend aktualisiert.

Plattform

Inferenz-Engine

Erste Schritte

Bereitstellen

sudo docker run --rm \
  --name cm5-hailo8-stdc1 \
  --privileged \
  --net=host \
  -e PYTHONUNBUFFERED=1 \
  --device /dev/hailo0:/dev/hailo0 \
  -v /usr/lib/libhailort.so.4.23.0:/usr/lib/libhailort.so.4.23.0:ro \
  -v /usr/lib/libhailort.so:/usr/lib/libhailort.so:ro \
  ghcr.io/seeed-projects/recomputer-hailo8-cv/stdc1:latest \
  python web_detection.py --model_path model/stdc1.hef --video_path video/test.mp4

REST API

Verwende die REST API für die Inferenz. Kopiere die folgenden Befehle.

Curl

curl -X POST "http://<Board_IP>:8000/api/models/stdc1/predict" \
  -F "file=@test.jpg"

Python

import requests

resp = requests.post(
    "http://<Board_IP>:8000/api/models/stdc1/predict",
    files={"file": open("test.jpg", "rb")},
    timeout=30,
)
print(resp.json())

Modelldetails

reComputer R Series (CM5 + Hailo-8)

STDC1 on reComputer R Series (CM5 + Hailo-8)

STDC1 (Short-Term Dense Concatenate) performs real-time semantic segmentation with 19 Cityscapes classes on Hailo-8 through HailoRT.

This page targets reComputer R Series (CM5 + Hailo-8) with a PCIe Hailo-8 accelerator.

Model Info

Property	Value
Architecture	STDC1 (Short-Term Dense Concatenate)
Task	Semantic segmentation
Input	1024x1920x3 RGB
Output	1024x1920 class mask (19 classes)
Parameters	8.27M
mIoU (Cityscapes)	73.8% hardware reference
HEF	Hailo Model Zoo v2.19.0, Hailo-8

The accuracy value above is Hailo's Model Zoo reference. It is not a benchmark measured on CM5.

Cityscapes Classes

ID	Class	ID	Class	ID	Class
0	road	7	traffic sign	14	truck
1	sidewalk	8	vegetation	15	bus
2	building	9	terrain	16	train
3	wall	10	sky	17	motorcycle
4	fence	11	person	18	bicycle
5	pole	12	rider
6	traffic light	13	car

Hardware and Host Setup

Item	Value
Board	reComputer R Series with Raspberry Pi CM5
Accelerator	Hailo-8 over PCIe, exposed as `/dev/hailo0`
Runtime	HailoRT 4.23.x
Python in container	3.11, aarch64

Install the Hailo-8 packages on the host and confirm that the driver and device are ready before starting the container:

bash

sudo apt update
sudo apt install hailo-all
sudo reboot

# After reboot
hailortcli --version
hailortcli fw-control identify
ls -l /dev/hailo0

The expected HailoRT version for this image is 4.23.x.

Run With Demo Video

bash

sudo docker run --rm \
  --name cm5-hailo8-stdc1 \
  --privileged \
  --net=host \
  -e PYTHONUNBUFFERED=1 \
  --device /dev/hailo0:/dev/hailo0 \
  -v /usr/lib/libhailort.so.4.23.0:/usr/lib/libhailort.so.4.23.0:ro \
  -v /usr/lib/libhailort.so:/usr/lib/libhailort.so:ro \
  ghcr.io/seeed-projects/recomputer-hailo8-cv/stdc1:latest \
  python web_detection.py --model_path model/stdc1.hef --video_path video/test.mp4

Open http://<Board_IP>:8000 to view the web preview.

USB Camera Mode

bash

sudo docker run --rm \
  --name cm5-hailo8-stdc1 \
  --privileged \
  --net=host \
  -e PYTHONUNBUFFERED=1 \
  --device /dev/hailo0:/dev/hailo0 \
  --device /dev/video0:/dev/video0 \
  -v /usr/lib/libhailort.so.4.23.0:/usr/lib/libhailort.so.4.23.0:ro \
  -v /usr/lib/libhailort.so:/usr/lib/libhailort.so:ro \
  ghcr.io/seeed-projects/recomputer-hailo8-cv/stdc1:latest \
  python web_detection.py --model_path model/stdc1.hef --camera_id 0

REST API

Prediction endpoint:

text

POST http://<Board_IP>:8000/api/models/stdc1/predict

Example image request:

bash

curl -X POST "http://<Board_IP>:8000/api/models/stdc1/predict" \
  -F "file=@test.jpg"

Endpoint	Method	Purpose
`/`	GET	Web preview UI
`/api/models/stdc1/predict`	POST	Segmentation mask (JSON)
`/api/models/stdc1/visualize`	POST	Overlay image (JPEG)
`/api/models/stdc1/classes`	GET	Cityscapes class list
`/api/video_feed`	GET	MJPEG preview stream

Development Notes

Source module: src/rpi5_hailo8_stdc1/
Dockerfile: docker/hailo8/stdc1.dockerfile
Container: ghcr.io/seeed-projects/recomputer-hailo8-cv/stdc1:latest
The module uses the HailoRT 4.23 VStreams API and a Hailo-8-specific HEF.

Eingaben und Ausgaben

Eingabe: Bild (1024x1920 RGB). Ausgabe: 19-Klassen semantische Segmentierungsmaske (Cityscapes).