Vision Observation Ingestion Service

This project is a FastAPI server for ingesting and processing observations from a computer vision system. It uses YOLO for object detection and maintains a "world state" based on the ingested observations.

Installation

1. Clone the repository:

   git clone <repository-url>
   cd <repository-directory>

2. Install the dependencies:

   pip install -r requirements.txt

Usage

To start the server, run the following command:

bash start.sh

The server will be available at http://localhost:8000.

API Reference

POST /observations

Ingests a new observation and updates the world state.

Request Body:

The request body should be a JSON object with the following structure:

{
  "source": "string",
  "timestamp": "datetime",
  "summary": "string",
  "objects": [
    {
      "label": "string",
      "confidence": "float"
    }
  ],
  "vision": {
    "key": "value"
  }
}

Response:

{
  "status": "ok"
}

GET /world

Returns the current world state.

Response:

A JSON object representing the current state of the world, as determined by the ingested observations.

GET /last_observation

Returns the last observation that was ingested.

Response:

A JSON object representing the last observation.