Skip to main content
FeaturesPricingDownloadAffiliate
Dashboard

Video Types

Boosta's AI uses different composition strategies based on the video type you select. Choosing the correct type ensures the framing and cropping logic matches your content.

Conversation​

Best for: Podcasts, Interviews, Panel Discussions

Backend Logic:

  • Priority: Multiple people.
  • Layout: Automatically switches between Single Speaker (focus on who is talking) and Split Screen (when two people speak or interact).
  • Distance Check: If speakers are close together, it uses a Wide Shot instead of splitting.
  • No Webcam: Does not look for picture-in-picture overlays.

Solo Creator​

Best for: Expert commentary, Talking heads, Tutorials

Backend Logic:

  • Priority: Single person.
  • Tracking: Smoothly tracks the speaker while keeping them centered.
  • Screen Detection: If the speaker shares their screen, the AI automatically switches to Screen & Reacts mode to ensure the content is visible.
  • Hook Detection: Analyzes speech patterns to identify engaging "hook" moments.

Screen & Reacts​

Best for: Gaming, Reaction videos, Software tutorials

Backend Logic:

  • Priority: Screen content.
  • Webcam: Aggressively searches for a facecam/webcam overlay.
  • Composition: Places the webcam at the bottom or corner of the vertical frame, leaving the top/center free for the main screen content.
  • Fallback: If no webcam is found, it reverts to Solo mode behavior.

Vlog / IRL​

Best for: Travel vlogs, Daily life, Street interviews

Backend Logic:

  • Priority: Group aware.
  • Fit Check: The AI calculates if multiple people fit within a single vertical crop.
    • If they fit: Uses a Group Shot to keep everyone in frame without splitting.
    • If they don't fit: Follows the Active Speaker.
  • Split Screen: Only used as a last resort if indistinguishable overlapping speech occurs.
  • Stabilization: Uses "stable crop" to smooth out handheld camera shake.

Faceless Content​

Best for: Documentaries, Stock footage narrations, Screen recordings (no face)

Backend Logic:

  • Priority: Visual content.
  • Face Tracking: Disabled. The AI ignores faces to prevent unnecessary zooming or jumping.
  • Smart Crop: Identifies key visual points of interest (text, objects, action) to center the crop.
  • Center Fallback: Defaults to a balanced center crop if no specific focal point is found.

Movies / TV Series​

Best for: Film clips, Trailers, Series highlights

Backend Logic:

  • Priority: Cinematic integrity.
  • Framing: Preserves Full Frame. The AI ensures the original composition is respected, often using a wider vertical crop (letterboxed) rather than zooming in and cutting off details.
  • Safety: No Split Screen. Never splits the screen to maintain the cinematic atmosphere.
  • Letterbox Removal: Automatically detects and removes hard-coded black bars from source footage before processing.