Detect people, cars, pets, and other objects in live camera frames using a Google Coral Edge TPU through Windows WSL2. Choose from 80 object categories and adjust detection confidence, processing rate, and input resolution.
中文简介正在准备中。当前显示英语简介。
Google Coral Edge TPU — real-time object detection natively via Windows WSL
安装前可以查看交给 AI 代理的指令内容。
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.还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Manage labeled datasets in Aegis Annotation Studio, including creating and deleting datasets, saving and loading annotations, and organizing labels. Export annotations in COCO format, a standard dataset structure, or upload datasets to Kaggle.
中文简介正在准备中。当前显示英语简介。
Connect Eufy cameras to SharpAI Aegis, discover cameras on the network, access live video through RTSP, a streaming protocol, and download event-triggered clips for analysis.
中文简介正在准备中。当前显示英语简介。
Connect Reolink cameras through their HTTP API and RTSP video streams. Capture clips when motion occurs, take snapshots, discover cameras through ONVIF, a camera interoperability standard, and obtain stream URLs for go2rtc live viewing.
中文简介正在准备中。当前显示英语简介。
Connect TP-Link Tapo cameras to discover devices on the network, capture motion-triggered clips, and obtain RTSP video stream URLs for live viewing through go2rtc, using RTSP and ONVIF camera integration.
中文简介正在准备中。当前显示英语简介。
Connect the Clawdbot agent in SharpAI Aegis to LINE Messenger to send messages and files, and reply to incoming messages through a messaging channel configured with LINE credentials.
中文简介正在准备中。当前显示英语简介。
Connect SharpAI Aegis's Clawdbot agent to a Matrix/Element room to search clips, receive alerts, and control cameras through conversation, with support for files, reactions, message edits, and threaded replies.
中文简介正在准备中。当前显示英语简介。