0:00 / 0:00

Mattia Gregoroni

@mattiamediax

INSTAGRAM · 2026-03-12Source

89likes

104comments

Remix This

Recreate with Kling 3 Motion Control

Make your own AI viral video

Prompt

GLOBAL LOCK: 
The video must maintain a consistent indoor studio environment with a muted green wall and a white minimalist shelf featuring small decorative objects (red house, framed photos). Lighting is soft, diffused front-daylight with natural highlight rolloff. The camera is a static medium close-up (MCU) at eye level, mimicking a 35mm lens with a shallow depth of field. All characters must follow the exact motion, head tilts, and facial expressions of the source reference video provided in the top-left corner. Speech is absent, but facial muscle movements must sync with the source's expressions.

[00:00–00:01]
Subject: A young African woman with deep dark skin, long straight dark hair with bangs, and striking emerald green eyes. She wears a glossy black vinyl sleeveless top.
Action: She looks directly into the lens, performing a subtle head tilt and a slight smile, mirroring the source video.
Lighting: Soft front light creating a luminous glow on her skin.

[00:01–00:03]
Subject: A man resembling Bad Bunny, with a short curly beard and hair. He wears a white dress shirt, a light-colored tie, and thin suspenders. He has small red-tinted rectangular sunglasses.
Action: He adjusts his sunglasses with his right hand while tilting his head slightly, perfectly synced to the source motion.
Lighting: Consistent studio lighting, sharp reflections on the sunglasses.

[00:03–00:04]
Subject: A man resembling Tom Hanks, middle-aged with short brown hair and blue eyes. He wears a beige blazer over a light blue shirt and tie.
Action: He has a slightly surprised or concerned expression, with raised eyebrows and a small mouth movement, following the source.

[00:04–00:05]
Subject: A man resembling Robert Downey Jr. as Tony Stark, wearing the Iron Man Mark 85 suit. The arc reactor in the center of the chest glows with a bright blue light.
Action: He raises his right hand, clenching it into a fist, while looking intensely at the camera. The metallic textures of the suit reflect the room's lighting.

[00:05–00:06]
Subject: A young blonde woman with long wavy hair and blue eyes. She wears a light blue sequined or textured top.
Action: She maintains a neutral but confident expression, looking directly forward as the camera focuses on her facial details.

[00:06–00:15]
Visual: Transition to a UI walkthrough of the Kling AI mobile/web application. The screen shows the "Motion Control" interface, selecting a character image of the African woman and mapping it to a source video of a man in a beige sweater.
Action: The cursor/finger navigates through "Add character image," "Motion Control," and "Bind Facial Element" settings.

[00:16–00:22]
Subject: The African woman from the first segment.
Action: She performs a complex series of gestures, including raising her hand to her chin and looking down then back up, perfectly mimicking the man in the source video overlay.
Text Overlay: "comment 'MOTION'" appears in bold white text, followed by "Save for later, Follow for more" on a grey background.

NEGATIVE PROMPT: 
Visual: robotic movement, distorted facial features, melting hands, extra fingers, flickering background, inconsistent lighting, blurry textures, uncanny valley eyes, warped clothing, text/logos on clothing.
Speech/Audio: (N/A as no speech is present, but avoid unnatural mouth warping).

SPEECH PACK:
(No speech present in this video. The audio is a rhythmic Latin music track).
Transcript: [Music playing: Bad Bunny - Titi Me Pregunto]
Sync Notes: Cuts between characters should land exactly on the downbeat of the music (approx. every 1-2 seconds).

How mattiamediax Made This Kling AI 3 Motion Control AI Video — and How to Recreate It

This case study analyzes a high-impact tech demonstration by @mattiamediax showcasing the capabilities of Kling AI 3.0's Motion Control feature. The video utilizes a "split-screen" comparison technique to demonstrate how AI can map human movement onto diverse digital avatars—ranging from hyper-realistic editorial portraits to world-famous celebrities like Bad Bunny and Tom Hanks. With a cinematic aesthetic characterized by warm, diffused indoor lighting, sharp 35mm-style depth of field, and flawless character consistency, this video serves as a masterclass in "Feature-First" content. It bridges the gap between complex AI technology and accessible creative output, making it highly shareable for indie creators looking to level up their production value without a Hollywood budget.

What You’re Seeing: A Visual Breakdown

The video is a fast-paced montage that transitions from high-fidelity AI results to a "behind-the-scenes" UI walkthrough. The primary subject is a series of characters—an African woman in a sleek black vinyl top, Bad Bunny in his signature red shades, Tom Hanks in a beige suit, and Robert Downey Jr. as Iron Man. Each character performs subtle head tilts, hand gestures, and facial expressions that perfectly mirror a source video of the creator. The environment remains a consistent, modern studio backdrop with a muted green wall and minimalist shelving, which helps maintain visual continuity across different character swaps.

Shot-by-Shot Breakdown

Time Range	Visual Content	Shot Language	Lighting & Color	Viewer Intent
00:00–00:01	African woman in black vinyl top mimicking a head tilt.	Medium Close-up (MCU), Static	Soft front-lit, high contrast	Hook: Immediate visual quality proof.
00:01–00:03	Bad Bunny character adjusting sunglasses.	MCU, 35mm feel	Warm tones, realistic skin	Celebrity recognition & "cool" factor.
00:03–00:04	Tom Hanks character with a concerned expression.	MCU, Eye-level	Natural daylight	Demonstrating emotional range.
00:04–00:05	Iron Man (RDJ) clenching a glowing fist.	MCU, Low angle feel	Metallic reflections, blue glow	Showcasing complex texture handling.
00:06–00:15	Kling AI App UI walkthrough (Motion Control).	Screen Recording	Dark Mode UI	Educational: "How it's done" reveal.
00:16–00:22	Final result: African woman mimicking complex hand gestures.	MCU, Dynamic motion	Consistent with start	Reinforce value & CTA.

Why It Went Viral: The Mechanics of AI Hype

The Power of "Celebrity Mimicry"

The choice of subjects is the primary driver of engagement. By using Bad Bunny and Iron Man, the creator taps into massive existing fanbases. Seeing a recognizable face perform a specific, controlled action (like Bad Bunny adjusting his glasses) triggers a "uncanny valley" fascination that keeps viewers watching. It’s not just a generic AI face; it’s a face the audience already has an emotional connection to, performing a task that looks impossibly real.

The "Secret Sauce" Reveal

The video follows a classic Result -> Process -> Result loop. By showing the UI of Kling AI 3.0 in the middle, the creator satisfies the viewer's curiosity ("How did they do that?") immediately after peaking their interest. This reduces the "barrier to entry" for the viewer, making them feel like they could also achieve these results, which drives saves and shares to "try this later."

Platform Signal Analysis

From a platform perspective, the 0–3 second hook is incredibly strong because it features a high-contrast, beautiful subject with a "Kling 3.0" watermark, signaling a new tech update. The loop effect is achieved by the rhythmic music (Bad Bunny's "Titi Me Pregunto") which matches the cuts, encouraging repeat views to catch the details of each character's movement. The "Comment 'MOTION'" CTA is a high-conversion tactic that boosts the engagement rate, signaling to the algorithm that the content is highly interactive.

5 Testable Viral Hypotheses

The Celebrity Anchor: Using a trending celebrity (Bad Bunny) in the first 3 seconds increases retention by 40% compared to generic avatars.
The UI Transparency: Showing the actual app interface builds trust and increases "Save" rates as it transforms "magic" into a "tutorial."
The Contrast Hook: Switching from a photorealistic woman to a stylized superhero (Iron Man) within 5 seconds prevents "scroll-past" by introducing visual variety.
The Keyword Trigger: Using a specific word like "MOTION" in the CTA allows for automated DM responses (via ManyChat), which exponentially increases engagement signals.
The Audio-Visual Sync: Cutting the character transitions exactly to the beat of a trending Latin track creates a "satisfying" viewing experience that encourages re-watching.

How to Recreate: From 0 to 1

Step 1: Topic Selection & Positioning

This format works best for "Tech-Creator" or "AI-Artist" accounts. Position yourself as the bridge between new tools and creative results.

Step 2: Source Video Preparation

Record a 5-10 second video of yourself performing clear, distinct movements (head tilts, hand waves, eye blinks). Ensure you are well-lit with a plain background to make the AI's job easier.

Step 3: Character Image Generation

Use Midjourney or Kling's internal image generator to create high-quality, front-facing portraits. Use prompts like: "Cinematic portrait of [Subject], 8k, highly detailed skin, studio lighting, looking at camera."

Step 4: Kling AI Motion Control

Upload your character image to Kling AI 3.0. Select the "Motion Control" or "Video-to-Video" feature. Upload your source video as the motion reference.

Step 5: Consistency Settings

Enable "Bind Facial Element" or "Character Consistency" toggles. This ensures the AI doesn't warp the face too much while trying to follow your movements.

Step 6: Generation & Iteration

Generate a 5-second clip. If the hands look "melty," reduce the motion strength or simplify your source video gestures.

Step 7: Editing & Pacing

In CapCut, sync your AI clips to a high-energy, rhythmic track. Place the "source video" in a small overlay window (Picture-in-Picture) to prove it's a mimicry effect.

Step 8: Publishing Strategy

Use a "Comment for Prompt" strategy. Tell users to comment a specific word to receive the exact settings you used. This explodes your comment count.

Growth Playbook: Distribution & Scaling

3 Ready-to-Use Opening Hooks

"Kling 3.0 just killed traditional video editing. Here’s why."
"I turned myself into Bad Bunny using only one AI tool."
"Stop scrolling! This is the most realistic AI video I’ve ever seen."

4 Caption Templates

The "New Update" Style: "Kling 3.0 is officially here and the Motion Control is 🤯. I tested it on 5 different characters and the results are scary. Which one looks most real? Comment 'AI' for the tutorial! 👇"
The "How-To" Style: "How to make any AI character move like you. 🧵 1. Record motion. 2. Upload to Kling. 3. Profit. Full breakdown in my bio! Save this for your next project."
The "Comparison" Style: "AI vs. Reality. 🤖 The gap is closing faster than we thought. I used Kling 3.0 to map my movements onto these avatars. Thoughts? Let's discuss in the comments."
The "Short & Punchy": "The future of content creation is here. Comment 'MOTION' to get the link to this tool! 🚀"

Hashtag Strategy

Broad (Reach): #AI #Tech #Innovation #Future #VideoProduction
Mid-Tier (Niche): #KlingAI #AIVideo #ContentCreatorTips #DigitalAvatar
Niche (Long-tail): #Kling30 #MotionControlAI #AICharacterConsistency #IndieCreatorTools

Frequently Asked Questions

What tools make it look the most similar?

Kling AI 3.0 is currently the leader for this specific "Motion Control" fidelity.

What are the 3 most important words in the prompt?

"Photorealistic," "Character Consistency," and "High-Fidelity Skin."

Why does the generated face look inconsistent?

Usually due to low-quality source images or too much movement in the reference video.

How can I avoid making it look like AI?

Use "35mm lens" and "natural lighting" in your image prompts to avoid the "plastic" AI look.

Is it easier to go viral on Instagram or TikTok with this?

Instagram Reels currently favors high-aesthetic tech demos like this one.

How should I properly disclose AI use?

Use the built-in "AI-Generated" labels on platforms and mention the tool in the caption.