Revolutionize Video Editing with TokenFlow Cognitive Actions

22 Apr 2025
Revolutionize Video Editing with TokenFlow Cognitive Actions

In the world of video processing, the ability to edit and enhance videos based on text prompts can significantly streamline workflows for developers and content creators alike. The TokenFlow platform offers an innovative Cognitive Action that leverages a pre-trained text-to-image diffusion model to perform consistent, text-driven video editing. This capability allows users to refine existing video content while maintaining spatial dynamics, ultimately producing high-quality results without the need for extensive training or fine-tuning.

Prerequisites

Before diving into the TokenFlow Cognitive Actions, ensure you have:

  • An API key for the Cognitive Actions platform to authenticate your requests.
  • Familiarity with JSON format for structuring input data.
  • A basic understanding of how to make HTTP requests from your application.

To authenticate your requests, you will typically pass your API key in the headers of your requests.

Cognitive Actions Overview

Edit Video with TokenFlow

The Edit Video with TokenFlow action allows users to edit videos based on descriptive text prompts, enhancing them using advanced diffusion techniques. This action falls under the category of video-processing.

Input

The input schema for this action requires a few mandatory fields and several optional parameters to fine-tune the video editing process:

{
  "video": "https://replicate.delivery/pbxt/KGcnUKDlfys2uAo8eHbWSE8Pzc17hI5IOGdstViaBvfPJWHY/wolf.mp4",
  "diffusionPrompt": "a shiny silver robotic wolf",
  "fps": 20,
  "width": 512,
  "height": 512,
  "guidanceScale": 7.5,
  "numberOfFrames": 40,
  "inversionPrompt": "",
  "numberOfDiffusionSteps": 50,
  "numberOfInversionSteps": 50,
  "negativeDiffusionPrompt": "ugly, blurry, low res, unrealistic, unaesthetic"
}
  • Required Fields:
    • video: The URL of the input video.
    • diffusionPrompt: A descriptive text that guides the video generation.
  • Optional Fields:
    • fps: Frames per second (default is 10).
    • seed: Seed for random value generation.
    • width and height: Dimensions of the output video (default is 512x512).
    • guidanceScale: Adjusts guidance levels (default is 7.5).
    • numberOfFrames: Total frames to process (default is 40).
    • inversionPrompt: Description of the input video (optional).
    • numberOfDiffusionSteps: Steps in the diffusion process (default is 50).
    • numberOfInversionSteps: Steps in the inversion process (default is 50).
    • negativeDiffusionPrompt: Characteristics to avoid in the output.

Example Input

Here’s a practical example of the JSON payload required to invoke the action:

{
  "fps": 20,
  "video": "https://replicate.delivery/pbxt/KGcnUKDlfys2uAo8eHbWSE8Pzc17hI5IOGdstViaBvfPJWHY/wolf.mp4",
  "width": 512,
  "height": 512,
  "guidanceScale": 7.5,
  "numberOfFrames": 40,
  "diffusionPrompt": "a shiny silver robotic wolf",
  "inversionPrompt": "",
  "numberOfDiffusionSteps": 50,
  "numberOfInversionSteps": 50,
  "negativeDiffusionPrompt": "ugly, blurry, low res, unrealistic, unaesthetic"
}

Output

Upon successful execution, the action typically returns a URL link to the processed video. For example:

https://assets.cognitiveactions.com/invocations/15170a0d-910b-465f-84bf-8402807216d0/620a52eb-c781-4f65-8d4c-c41a3e9d5cdc.mp4

This output URL points to the newly edited video, ready for further use or distribution.

Conceptual Usage Example (Python)

Here’s a conceptual Python code snippet illustrating how to call the TokenFlow Cognitive Action:

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 = "99ad2556-3944-436e-8547-cbb03eee99cb" # Action ID for Edit Video with TokenFlow

# Construct the input payload based on the action's requirements
payload = {
    "fps": 20,
    "video": "https://replicate.delivery/pbxt/KGcnUKDlfys2uAo8eHbWSE8Pzc17hI5IOGdstViaBvfPJWHY/wolf.mp4",
    "width": 512,
    "height": 512,
    "guidanceScale": 7.5,
    "numberOfFrames": 40,
    "diffusionPrompt": "a shiny silver robotic wolf",
    "inversionPrompt": "",
    "numberOfDiffusionSteps": 50,
    "numberOfInversionSteps": 50,
    "negativeDiffusionPrompt": "ugly, blurry, low res, unrealistic, unaesthetic"
}

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 example, make sure to replace the placeholder API key and endpoint with actual values. The action_id and payload are structured as per the action requirements.

Conclusion

The TokenFlow Cognitive Action for video editing presents a powerful tool for developers looking to enhance video content through innovative, text-driven techniques. By integrating this action into your applications, you can unlock new creative possibilities and streamline video production workflows. Consider exploring additional use cases, such as automated video enhancement, custom video generation for marketing, or interactive media applications.

← Back to All Articles