8 Practical Google Gemini Prompts for Video Editing and Short-Form Content Creation

As Google Gemini's multimodal video understanding and generative editing capabilities expand, prompt-driven video editing is emerging as a compelling alternative to manual timeline adjustments in traditional non-linear editors and mobile apps. AI creator @naiem_bn shared a collection of 8 practical video editing prompt templates on X, designed to execute precise edits on existing footage while strictly preserving subject identity, voice synchronization, and real-world physical consistency.

Image source: @naiem_bn via X

These templates use bracketed placeholders ([...]) where creators can inject their specific parameters and visual requirements. Below is a detailed breakdown of all 8 prompts, optimized for creator workflows and social video production.

1. Turn a Basic Recording into a Professional Reel

Transforms a casual talking-head recording filmed on a smartphone or webcam into a studio-grade Instagram Reel or YouTube Short, with upgraded lighting, background isolation, and social-first framing.

Edit this uploaded talking-head video into a polished, professional social media Reel. Preserve my exact identity, facial features, voice, spoken words, lip-sync, expressions, and natural body movements. Replace the background with [BACKGROUND]. Improve the lighting with a soft, realistic studio setup, add subtle cinematic color grading, and keep my face sharp and naturally detailed. Reframe the composition for 9:16 vertical format, keep me centered, and leave clean negative space for captions. Make the final result look professionally filmed—not overly edited or artificial.

• Key Parameters: [BACKGROUND] (e.g., modern minimalist studio, warm home library, ambient workspace).
• How It Works: Locks facial geometry, voice track, and lip-sync while replacing the background environment. It reframes the shot into 9:16 vertical format, centering the speaker and maintaining negative margin space for caption overlays.

2. Convert a Horizontal Video into a Vertical Short

Intelligently converts 16:9 landscape footage into a 9:16 vertical video without simply chopping off the left and right sides of the frame.

Convert this landscape video into a professional 9:16 vertical video optimized for [INSTAGRAM REELS/TIKTOK/YOUTUBE SHORTS]. Keep [MAIN SUBJECT] clearly visible throughout the entire clip. Reframe intelligently instead of simply cropping important content, and reconstruct missing areas naturally when necessary. Preserve the original faces, movements, audio, timing, expressions, and visual style. Keep important text, products, and subjects inside the central safe zone. Make the final framing feel intentional and professionally composed.

• Key Parameters: Target platform (Instagram Reels / TikTok / YouTube Shorts), [MAIN SUBJECT].
• How It Works: Reconstructs out-of-frame upper and lower areas via inpainting while keeping focal subjects and key products inside the UI safe zone, producing a naturally framed vertical composition rather than an aggressive crop.

3. Create a Cinematic Product Advertisement

Upgrades raw product footage into a high-end commercial asset with dynamic camera movement, environmental styling, and realistic product effects.

Transform this product footage into a premium cinematic advertisement for [PRODUCT NAME]. Preserve the product’s exact shape, proportions, colors, packaging, logo, typography, and label details without distortion. Place the product in [SETTING] with [LIGHTING STYLE]. Create a smooth cinematic camera movement from [STARTING ANGLE] to [ENDING ANGLE]. Add subtle realistic [PARTICLES/STEAM/WATER DROPLETS] that complement the scene without covering the product. Build toward a powerful hero shot and finish with the product perfectly framed. Use a [LUXURY/ENERGETIC/MINIMAL] visual direction with premium commercial quality.

• Key Parameters: [PRODUCT NAME], [SETTING], [LIGHTING STYLE], [STARTING ANGLE] to [ENDING ANGLE], practical elements (steam, water droplets, dust particles), aesthetic tone.
• How It Works: Strictly preserves brand logos, typography, and packaging geometry from hallucination while applying procedural camera sweeps and atmospheric accents that culminate in a hero product shot.

4. Replace a Boring Background

Swaps out an uninspiring or messy room behind a subject for a completely new location while preventing green-screen fringing, mismatched lighting angles, or plastic cutout borders.

Replace the existing background of this video with [NEW LOCATION] while keeping the main subject completely unchanged. Preserve the person’s exact identity, face, hair, clothing, body proportions, movements, voice, and lip-sync. Match the new environment to the original camera angle, perspective, lighting direction, shadows, reflections, depth of field, and color temperature. Add subtle environmental motion such as [MOVING LIGHTS/TREES SWAYING/PEOPLE WALKING] so the subject looks naturally filmed inside the new location. Avoid obvious cutout edges, artificial lighting, or compositing artifacts.

• Key Parameters: [NEW LOCATION], ambient motion cues (swaying trees, background pedestrians, soft traffic lights).
• How It Works: Harmonizes the color temperature, lighting direction, and depth of field between the subject and the new environment, adding subtle dynamic elements in the backdrop to maintain scene realism.

5. Fix Poor Lighting and Colors

Rescues underexposed, muddy, or poorly color-balanced footage without introducing extreme digital noise or synthetic skin smoothing.

Professionally color-correct and enhance this poorly lit video without changing its original content. Fix exposure, white balance, contrast, and uneven lighting. Restore natural and accurate skin tones, recover shadow detail, soften harsh facial shadows, and reduce blown-out highlights. Add flattering directional light from [LEFT/RIGHT/FRONT] while maintaining realistic facial texture and natural skin detail. Apply a clean [WARM CINEMATIC/BRIGHT COMMERCIAL/COOL MODERN] color grade. Keep the result realistic and professionally shot—avoid plastic skin, excessive smoothing, oversaturation, or an artificial HDR look.

• Key Parameters: Direction of key light ([LEFT/RIGHT/FRONT]), grading aesthetic ([WARM CINEMATIC/BRIGHT COMMERCIAL/COOL MODERN]).
• How It Works: Balances exposure curves and white balance, lifts crushed shadows, and introduces virtual key lighting while retaining natural skin pores and realistic facial texture.

6. Remove Distracting Objects

A generative video inpainting prompt that cleans up cables, background clutter, photobombers, or unwanted props from an entire clip.

Remove [UNWANTED PERSON/OBJECT/CABLE/CLUTTER] from this entire video. Reconstruct the hidden background naturally and consistently across every frame. Preserve the original camera movement, perspective, lighting, shadows, reflections, textures, and surrounding objects. Do not alter the main subject, facial features, clothing, product details, dialogue, audio, timing, or composition. Ensure the removed area remains stable throughout the clip with no flickering, warping, ghosting, duplicated objects, or visible editing artifacts.

• Key Parameters: [UNWANTED PERSON/OBJECT/CABLE/CLUTTER].
• How It Works: Reconstructs occluded surfaces across each frame with temporal consistency, ensuring background textures, reflections, and shadow gradients stay locked without flickering or warping.

7. Add an Attention-Grabbing Visual Effect

Seamlessly integrates realistic cinematic visual effects into raw footage at an exact cue or physical action.

Add a realistic cinematic [GLOWING PORTAL/HOLOGRAM/ENERGY TRAIL/MAGICAL PARTICLES] precisely when [DESCRIBE THE EXACT ACTION OR MOMENT]. Make the effect interact naturally with the subject and environment. Match its perspective, scale, lighting, shadows, reflections, depth, motion blur, and camera movement to the original footage. Keep the person, face, clothing, environment, audio, and original camera motion unchanged. Make the effect visually impressive and believable while keeping it clean enough that it supports the story rather than distracting from the subject.

• Key Parameters: VFX type, trigger moment ([DESCRIBE THE EXACT ACTION OR MOMENT]).
• How It Works: Calculates camera motion blur, scene perspective, and subject lighting interaction so that simulated elements (energy trails, holographic displays, portals) blend convincingly into the physical space.

8. Build a Complete Social Video from One Topic

Generates an end-to-end short-form vertical video from a high-level concept, including an initial hook, structured pacing, B-roll transitions, voiceover, and call to action.

Create a complete [30/45/60]-second vertical social media video about [TOPIC] for [TARGET AUDIENCE]. Open with a powerful curiosity-driven hook within the first two seconds. Deliver three practical and valuable points using short, visually engaging scenes. Use a [FRIENDLY/EXPERT/HUMOROUS] voiceover, concise on-screen captions, relevant B-roll, smooth transitions, and [MUSIC STYLE]. Finish with this clear call to action: ‘[CTA].’ Keep the language simple, the pacing fast, and every scene visually distinct. Make the final video feel professionally produced, engaging, and ready to publish.

• Key Parameters: Clip duration (30/45/60 seconds), [TOPIC], [TARGET AUDIENCE], voiceover personality, [MUSIC STYLE], [CTA] (Call to Action).
• How It Works: Structures the video to capture retention within the first two seconds, presenting three distinct informational beats with synchronized voiceover and captions before concluding with an engagement prompt.

Original source

• @naiem_bn on X: https://x.com/naiem_bn/status/2107452507165434043