Unlocking Character and Style Consistency with Runway's Updated Video-to-Video
Instead of relying purely on text prompts (which often led to inconsistent visuals), you can now upload a restyled first frame. This means you get full freedom of movement (since your input is a video file) and total precision over your video’s visual style (since you're preparing this in your favourite image generator). Consistent characters, exact style, complete creative control.
Creating visually consistent, stylised videos has long been a challenge for filmmakers and creators experimenting with generative AI tools. Runway’s original Video-to-Video (V2V) feature was powerful, transforming videos into visually impressive scenes, but it had limitations:
Inconsistent Characters: Your character could look slightly different in each shot, disrupting continuity.
Visual Style Variations: Each frame was stylised uniquely, making it tough to stitch together multiple shots seamlessly.
Thankfully, Runway’s recent V2V update completely changes the game, enabling filmmakers to:
✅ Maintain the same character and visual style across multiple shots.
✅ Enjoy complete camera and character movement control.
Here's exactly how you can use this updated workflow in your own video projects:
Step-by-Step Workflow for Perfect Visual Consistency
01. Shoot Your Footage
Use your smartphone or camera to record your video. The better your footage, the cleaner the results.
02. Upload Your Video to Runway
Simply drop your footage into Runway’s Video-to-Video tool.
03. Grab the First Frame & Restyle It
Extract the first frame and restyle it in your favourite image-editing AI tool. I experimented with Leonardo.ai and Magnific AI for best results.
04. Re-upload the Restyled Frame to Runway
This stylised frame acts as a visual guide for Runway's AI, ensuring every subsequent frame aligns with this style.
05. Adjust Structure Transformation Settings
Play around to find the ideal setting. I found structure transformation level 4 offered the best balance of detail and consistency.
06. Repeat the Process for Each Shot
This ensures all your shots have the same style and character consistency.
07. Stitch Together in Editing Software
Use any editing software to connect your shots smoothly. Now you have a coherent, stylized video sequence.
Two Essential Tips for Optimal Results
🎬 Preparation is Everything
The original props, costumes, and environments heavily guide AI transformation accuracy. If you're drastically changing appearance, use wigs, costumes, or relevant props to help the AI understand your visual intent.
🗣️ Fix Lip-Sync Issues Easily
Significant style transformations might affect lip-sync. Thankfully, Runway's Act One solves this by easily syncing original lip movements back into your stylized video.
Creativity in Action
To illustrate this, I created a quick scene at home using nothing more elaborate than my kitchen scale as a futuristic prop. The results were impressive, and a reminder that creativity often involves simple tools combined with powerful tech.
Check out the example video to see how seamless and effective the updated V2V workflow can be.
Explore, experiment, and unlock your creativity with Runway's updated Video-to-Video feature. Happy filmmaking!
👍 Like this post? Share it with someone who'd benefit.
➕ Follow me on LinkedIn for more insights into GenAI filmmaking, practical workflows, and creative tips.