# DOODS

DOODS is a self-hosted, containerized REST service that detects objects in images and video streams, with optional GPU and EdgeTPU hardware acceleration. It is open source and self-hosted, installable on Docker Hub and Unraid.

Canonical page: https://ourlocalhost.com/app/1286/doods

## Description

DOODS (Dedicated Open Object Detection Service) is a self-hosted, containerized REST service that detects objects in images and video streams, with optional GPU and EdgeTPU hardware acceleration.

## Platform availability

Available on 1 of 9 platforms: Unraid.

## Health score

47/100 (Fair), based on 4 of 4 factors.
- Maintained: 61 (last commit 144 days ago)
- Popular: 49 (277 GitHub stars)
- Easy to install: 35 (1 install platform, install notes, Docker image)
- Light to run: 20 (8 GB minimum RAM, runs on ARM)

## Links

- Docker Hub: https://hub.docker.com/r/snowzach/doods2
- GitHub: https://github.com/snowzach/doods2
- Unraid: https://ca.unraid.net/apps/doods-0tr6sgr13a600f?q=DOODS

## Common questions

### Can DOODS2 use GPU or EdgeTPU acceleration?

Yes, DOODS2 supports GPUs and EdgeTPU hardware acceleration.

### How can DOODS be configured for Home Assistant?

It can run locally or remotely in a Docker container and is configured almost exactly like the Tensorflow component.

### Does DOODS2 remain compatible with the original DOODS REST API?

Yes, DOODS2 supports the same REST API endpoints as original DOODS. It also adds endpoints for streaming feeds with annotated video and websocket JSON detection data.

### What service interfaces does DOODS2 expose?

DOODS2 exposes REST and gRPC interfaces for object detection on images and streams.

### Is there a Hass.io add-on repository for DOODS2?

A Hass.IO repository for snowzach’s add-ons is available, but the provided source only points to the repository and does not describe setup steps.

---
Source: Our Local Host (https://ourlocalhost.com/app/1286/doods), the self-hosted and open-source app directory.