StreamDeploy Curated Containers
Fast-start containers for fleet-grade deployments: ROS 2 camera streaming, TensorRT inference, and Coral Edge TPU detection. Designed for NVIDIA Orin/Jetson and similar edge AI hardware. All containers support: DEPLOY_ENV, structured logs, healthcheck, and configurable camera/device sources. Use with StreamDeploy’s deployment pipeline for robots to get OTA updates, versioned configs, and one-command rollbacks.
Filters:
3 of 3 shown
ros2-vision-streamer
Multi-camera ROS 2 + GStreamer (Jetson/Orin-first)
Primary HW
NVIDIA Orin/Jetson (arm64)
What it does
Multi-camera capture, H.264/H.265 encode via GStreamer; publishes ROS 2 topics; optional RTSP.
arm64ROS 2GStreamerRTSPNVENC
triton-robot-inference
TensorRT/ONNX runtime FastAPI microservice (Orin-first)
Primary HW
NVIDIA Orin/Jetson (arm64)
What it does
Lightweight FastAPI microservice running ONNX Runtime/TensorRT for common vision models.
arm64FastAPIONNXTensorRTJetson
coral-rtsp-detector
Coral Edge TPU RTSP detector (Coral Dev Board/USB)
Primary HW
Coral Dev Board / USB Accelerator (arm64/amd64 host)
What it does
Runs TFLite Edge TPU detection and serves annotated RTSP stream.
arm64amd64CoralEdgeTPURTSPTFLite