image-animation

Animate Your Images with Lia's Latent Space Navigation

26 Apr 2025
Animate Your Images with Lia's Latent Space Navigation

In an era where visual content reigns supreme, the ability to bring static images to life has become increasingly valuable. Lia offers a powerful Cognitive Action that leverages latent space navigation to animate images, transforming them into engaging video content. This innovative approach utilizes advanced models trained on datasets like 'vox', 'taichi', and 'ted', allowing developers to create dynamic visual experiences with ease. Whether you're looking to enhance marketing materials, create captivating social media posts, or simply explore new artistic avenues, this action simplifies the animation process, making it accessible and efficient.

Prerequisites

To get started, you will need a Cognitive Actions API key and a basic understanding of how to make API calls.

Animate Images via Latent Space Navigation

This action, "Animate Images via Latent Space Navigation," allows developers to animate static images by employing models trained on specific datasets. The underlying technology is based on the PyTorch implementation of the ICLR 2022 paper titled 'Latent Image Animator'. By navigating the latent space, you can create visually stunning animations that breathe life into your images.

Input Requirements: To utilize this action, you will need to provide:

  • drivingVideo: A valid URI pointing to the driving video you wish to use. This video will dictate the movement and style of the animation.
  • imageSource: A valid URI pointing to the input source image that you want to animate.
  • model: A string indicating which dataset to use, with options including 'vox', 'taichi', or 'ted'. The default is 'vox'.

Example Input:

{
  "model": "vox",
  "imageSource": "https://replicate.delivery/mgxm/8140a5a9-16c5-42ee-90bb-cd755f8ca9bd/mona_lisa.png",
  "drivingVideo": "https://replicate.delivery/mgxm/1724727e-d9ea-449b-9cca-2c39a9004350/driving2.mp4"
}

Expected Output: The output will be a URI pointing to the animated video generated from the static image and the driving video, allowing you to easily access and share your creation.

Example Output:

https://assets.cognitiveactions.com/invocations/69b2c34c-8575-416e-aa71-35a8826b57b0/ea35347b-a52f-4752-9b2c-057ef6ce0e49.mp4

Use Cases for this Specific Action:

  • Marketing and Advertising: Create eye-catching advertisements that capture viewer attention by animating product images.
  • Social Media Content: Enhance your social media presence with animated images that stand out in feeds, driving engagement and shares.
  • Art and Creativity: Explore new artistic expressions by animating classic artworks or personal photographs, providing a modern twist on traditional imagery.
  • Education and Presentations: Use animated images to make educational content more engaging and visually appealing, helping to retain audience interest.
import requests
import json

# Replace with your actual Cognitive Actions API key and endpoint
# Ensure your environment securely handles the API key
COGNITIVE_ACTIONS_API_KEY = "YOUR_COGNITIVE_ACTIONS_API_KEY"
# This endpoint URL is hypothetical and should be documented for users
COGNITIVE_ACTIONS_EXECUTE_URL = "https://api.cognitiveactions.com/actions/execute"

action_id = "b9e98dfd-9523-4ddd-a5c2-b6e45ac19817" # Action ID for: Animate Images via Latent Space Navigation

# Construct the exact input payload based on the action's requirements
# This example uses the predefined example_input for this action:
payload = {
  "model": "vox",
  "imageSource": "https://replicate.delivery/mgxm/8140a5a9-16c5-42ee-90bb-cd755f8ca9bd/mona_lisa.png",
  "drivingVideo": "https://replicate.delivery/mgxm/1724727e-d9ea-449b-9cca-2c39a9004350/driving2.mp4"
}

headers = {
    "Authorization": f"Bearer {COGNITIVE_ACTIONS_API_KEY}",
    "Content-Type": "application/json",
    # Add any other required headers for the Cognitive Actions API
}

# Prepare the request body for the hypothetical execution endpoint
request_body = {
    "action_id": action_id,
    "inputs": payload
}

print(f"--- Calling Cognitive Action: {action.name or action_id} ---")
print(f"Endpoint: {COGNITIVE_ACTIONS_EXECUTE_URL}")
print(f"Action ID: {action_id}")
print("Payload being sent:")
print(json.dumps(request_body, indent=2))
print("------------------------------------------------")

try:
    response = requests.post(
        COGNITIVE_ACTIONS_EXECUTE_URL,
        headers=headers,
        json=request_body
    )
    response.raise_for_status() # Raise an exception for bad status codes (4xx or 5xx)

    result = response.json()
    print("Action executed successfully. Result:")
    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 (non-JSON): {e.response.text}")
    print("------------------------------------------------")

Conclusion

Lia's Cognitive Action for animating images via latent space navigation opens up a world of possibilities for developers looking to create dynamic visual content. By simplifying the animation process, this action allows you to focus on creativity and storytelling rather than technical complexities. Whether for marketing, social media, or artistic exploration, the potential applications are vast. Start integrating this action today and transform your static images into captivating animations that resonate with your audience.

← Back to All Articles