​
Node Inputs

​
Required Fields

  • Video File: Upload video (MP4/MOV/AVI)
  • Prompt: Question or instruction for analysis. Be specific for best results
  • Video Model: Choose AI model for analysis (Gemini 1.5 Flash/Pro)

​
Optional Fields

  • Use Link: Enable to use video URLs instead of files
  • Temperature: Controls analysis creativity (0-1, default: 1)
  • Cache Response: Save responses for reuse

​
Show As Input Options

You can expose these fields as inputs under the Configure Inputs option:

  • Prompt
  • Temperature

​
Node Output

  • Analysis: AI’s detailed response about the video content

​
Node Functionality

The Analyze Video node can:

  • Process video content with AI vision
  • Extract text from video frames
  • Generate video descriptions
  • Answer queries about video content
  • Identify objects and actions
  • Analyze scenes and transitions
  • Track movement and changes

​
Available AI Models

  • Gemini 1.5 Pro: Advanced analysis with higher accuracy
  • Gemini 1.5 Flash: Faster processing for simpler tasks

​
Common Use Cases

  1. Content Description:
Prompt: "Describe the main events in this video"
Use: Content cataloging, accessibility
  1. Text Extraction:
Prompt: "Extract any text that appears in the video"
Use: Subtitle extraction, text analysis
  1. Action Recognition:
Prompt: "List all activities and actions occurring in this video"
Use: Behavior analysis, event detection
  1. Object Tracking:
Prompt: "Track and describe the movement of specific objects"
Use: Motion analysis, surveillance

​
Best Practices

  1. Video Length:

    • Keep videos reasonably short for better analysis
    • Consider splitting longer videos into segments

  2. Prompt Engineering:

    • Be specific in your instructions
    • Focus on one analysis task at a time
    • Include temporal aspects if timing is important

  3. Model Selection:

    • Use Gemini 1.5 Pro for detailed analysis
    • Choose Gemini 1.5 Flash for quick, basic analysis

  4. Temperature Usage:

    • Low (0-0.3): Consistent, factual analysis
    • Medium (0.4-0.7): Balanced analysis
    • High (0.8-1.0): Creative descriptions

​
Important Considerations

  1. Cost: 20 credits per video analysis
  2. Processing time increases with video length
  3. Analysis quality depends on video resolution and clarity

In summary, the Analyze Video node provides powerful video analysis capabilities using state-of-the-art AI vision models, suitable for a wide range of applications from content description to detailed motion analysis.

Analyze ImageGenerate Image