Detect and Classify Images Effortlessly with meta/detic Cognitive Actions

24 Apr 2025
Detect and Classify Images Effortlessly with meta/detic Cognitive Actions

In the realm of image analysis, effectively detecting and classifying images has become increasingly vital. The meta/detic specification provides developers with powerful Cognitive Actions that harness advanced models to recognize objects in images with high accuracy. This guide will walk you through one of the key actions available, enabling you to integrate image classification capabilities into your applications seamlessly.

Prerequisites

Before you dive into using the Cognitive Actions from the meta/detic spec, ensure you have the following:

  • An API key for the Cognitive Actions platform.
  • Access to a valid endpoint for executing the actions.
  • Familiarity with JSON payloads, as you'll be working with them for input and output data.

Authentication

Authentication typically involves passing your API key in the headers of your requests. This ensures that your application is authorized to access the Cognitive Actions services.

Cognitive Actions Overview

Detect Image Classes with Detic

This action utilizes the Detic model to detect various classes specified by training on image-level labels. By leveraging the CLIP model, it provides state-of-the-art results on open-vocabulary datasets, allowing for cross-dataset generalization without the need for fine-tuning.

  • Category: Image Analysis

Input

The input schema for this action requires the following fields:

  • image (required): The URI of the input image. It must point to a valid image resource.
  • vocabularyType (optional): This field allows you to select the vocabulary type to use for detection. The options include:
    • lvis (default)
    • objects365
    • openimages
    • coco
    • custom
  • customVocabulary (optional): If the vocabulary type is set to custom, you can specify entries separated by commas.

Example Input:

{
  "image": "https://replicate.delivery/mgxm/d4825bcc-d07f-4c85-91d4-b3a5b14067aa/k67kjlB.jpeg",
  "vocabularyType": "lvis",
  "customVocabulary": ""
}

Output

The output of this action typically returns a URL pointing to an image that showcases the detected classes.

Example Output:

https://assets.cognitiveactions.com/invocations/b4db0fec-2d9e-4f2d-97c0-1112afab48cd/6f92ea34-74a3-480d-997f-cb56c4c4432b.png

Conceptual Usage Example (Python)

Here's how you might call this action using a hypothetical Cognitive Actions execution endpoint with Python:

import requests
import json

# Replace with your Cognitive Actions API key and endpoint
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"  # Hypothetical endpoint

action_id = "7a022bcb-e046-492d-971f-10f0efa7b9d5"  # Action ID for Detect Image Classes with Detic

# Construct the input payload based on the action's requirements
payload = {
    "image": "https://replicate.delivery/mgxm/d4825bcc-d07f-4c85-91d4-b3a5b14067aa/k67kjlB.jpeg",
    "vocabularyType": "lvis",
    "customVocabulary": ""
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json"
}

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json={"action_id": action_id, "inputs": payload}  # Hypothetical structure
    )
    response.raise_for_status()  # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully:")
    print(json.dumps(result, indent=2))

except requests.exceptions.RequestException as e:
    print(f"Error executing action {action_id}: {e}")
    if e.response is not None:
        print(f"Response status: {e.response.status_code}")
        try:
            print(f"Response body: {e.response.json()}")
        except json.JSONDecodeError:
            print(f"Response body: {e.response.text}")

In this snippet, you will need to replace YOUR_COGNITIVE_ACTIONS_API_KEY with your actual API key. The action_id corresponds to the action you are executing, and the payload is structured according to the required input schema. Note that the exact endpoint URL and request structure are illustrative.

Conclusion

The meta/detic Cognitive Actions provide developers with a streamlined approach to image classification and object detection. By utilizing the advanced capabilities of the Detic model, you can easily integrate powerful image analysis features into your applications. Explore the possibilities of enhancing your projects with these actions and consider how they can be tailored to meet your specific needs. Happy coding!

← Back to All Articles