A full-body deepfake manipulates body movement, posture, or full-figure footage rather than just the face. These attacks combine face swap, body re-rendering, and motion synthesis to fabricate complete physical performances.

• A neural model captures pose and motion from a source video
• The target’s appearance is rendered onto the captured pose with consistent clothing, hair, and lighting
• Recent models can generate plausible full-body video from a single reference image

• Fake protest or crowd footage used in disinformation campaigns
• Manipulated security camera footage submitted as legal evidence
• Fabricated athletic performance video for sports betting fraud
• Synthetic compromising footage for blackmail or harassment
• Fake influencer content showing celebrities in fabricated locations or activities