Google Coral Edge TPU — real-time object detection natively via Windows WSL
Real-time object detection natively utilizing the Google Coral Edge TPU accelerator on your local hardware via Windows Subsystem for Linux (WSL). Detects 80 COCO classes (person, car, dog, cat, etc.) with ~4ms inference on 320x320 input.
usbipd-win installed on the Windows host┌─────────────────────────────────────────────────────┐
│ Host (Aegis-AI on Windows) │
│ frame.jpg → /tmp/aegis_detection/ │
│ stdin ──→ ┌──────────────────────────────┐ │
│ │ WSL Container / Environment │ │
│ │ detect.py │ │
│ │ ├─ loads _edgetpu.tflite │ │
│ │ ├─ reads frame from disk │ │
│ │ └─ runs inference on TPU │ │
│ stdout ←── │ → JSONL detections │ │
│ └──────────────────────────────┘ │
│ USB ──→ usbipd-win bridge to WSL │
└─────────────────────────────────────────────────────┘
/tmp/aegis_detection/ workspaceframe event via stdin JSONL to the WSL Python instancedetect.py invokes PyCoral and executes natively on the mapped USB Edge TPU inside Linuxdetections event via stdout JSONL back to Windows Host| Input Size | Inference | On-chip | Notes |
|---|---|---|---|
| 320x320 | ~4ms | 100% | Fully on TPU, best for real-time |
| 640x640 | ~20ms | Partial | Some layers on CPU (model segmented) |
Cooling: The USB Accelerator aluminum case acts as a heatsink. If too hot to touch during continuous inference, it will thermal-throttle. Consider active cooling or
clock_speed: standard.
Run deploy.bat — this will:
usbipd is installed and bind the 18d1:9302 and 1a6e:089a Edge TPU hardware IDs.usbipd seamlessly during invocation.