- Higgsfield Katana templates are preset editing directions that pair a reference video with your own assets.
- Four core presets: Personalized Edits, Style Transfer, Motion Transfer, and Product/Social workflows.
- Best starting point: Personalized Edits for swapping subjects while keeping reference motion and pacing.
- Key habit: Use conversational follow-up prompts to refine results instead of rebuilding from scratch.
- Credit tip: Every rerun may consume additional credits—review the estimate before generating.
What Are Higgsfield Katana Templates?
Higgsfield Katana templates (also called presets) are structured starting workflows that tell the AI how to treat your reference video and uploaded assets. Instead of describing every creative decision from scratch, you select a preset that matches your goal—replacing a subject, transferring a style, or carrying motion into a new character—then supply the reference video, images, and a concise prompt.
The four preset families cover nearly every common short-form video scenario:
| Template | Category | Primary Goal | Best For |
|---|---|---|---|
| Personalized Edits | Subject Adaptation | Swap subject/product while keeping structure | Product ads, personalized social videos |
| Style Transfer | Visual Transformation | Change look, palette, atmosphere | Fashion films, brand experiments |
| Motion Transfer | Movement Adaptation | Carry performance to a new subject | Action clips, performance content |
| Product & Social Workflows | Content Scenario | Build short-form content around a format | E-commerce, campaigns, social hooks |
Choose the preset by what matters most in the reference: if the movement is irreplaceable, use Motion Transfer; if only the look needs to change, use Style Transfer; if you want to put your own subject into the structure, use Personalized Edits.
Template Comparison at a Glance
Each template has different asset requirements and iteration behavior. This comparison helps you match a template to your available material before you spend credits.
| Factor | Personalized Edits | Style Transfer | Motion Transfer | Product & Social |
|---|---|---|---|---|
| Reference video needed | Yes | Yes | Yes, with readable motion | Yes, format-matched |
| Character image | Optional | Optional | Yes | Optional |
| Product image | Yes (for product goals) | No | No | Yes |
| Clothing reference | Optional | Optional | Optional | Optional |
| Prompt focus | What stays vs. changes | Color, lighting, mood | Subject-motion relationship | Format and emphasis |
| Iteration style | Targeted swaps | Style-only edits | Motion checks first | Hook variations |
Personalized Edits
- Best all-rounder for creators with their own subject or product
- Preserves camera rhythm and pacing from the reference
- Requires a clear description of what changes and what stays
Style Transfer
- Ideal when the original movement is valuable but the visual identity must change
- Define style through color, lighting, atmosphere, and materials
- Keep style descriptions consistent across the whole sequence
Motion Transfer
- Prioritizes the motion language of the reference
- Works best with one readable action, not rapid cuts
- Check subject visibility and continuity in every output
Product & Social Workflows
- Built around short-form formats and social hooks
- Combines product, wardrobe, or character assets with a concise prompt
- Generate multiple targeted variations while watching credit use
If your character, product, and clothing images describe different intended subjects, the generated video will drift from the reference. Supply consistent references that all point to the same visual target.
Step-by-Step: Using a Katana Template
Follow this workflow to go from a reference clip to a refined generated result. The same six steps apply to every template—only the emphasis of your prompt changes.
Choose a Reference Video
Pick footage with clear subject movement and readable camera actions. Match the reference type to your goal: product ad, fashion clip, action sequence, or social video. Avoid heavy obstruction and unnecessary cuts.
Upload Your Creative Assets
Add character, product, or clothing images that show the subject from useful angles. Describe each asset's role so Katana can distinguish the replacement subject from the reference footage.
Select the Template Direction
Choose Personalized Edits, Style Transfer, or Motion Transfer based on the comparison above. Combine directions only when your prompt clearly separates each goal.
Describe the Intended Changes
State what must remain consistent (camera rhythm, main action) and what should change (subject, product, palette, wardrobe, environment). Keep instructions concise and non-contradictory.
Generate and Review
Check whether the main action, framing, and pacing match the reference. Verify consistency in the replacement subject, product details, and overall style before iterating.
Refine with a Follow-Up Edit
Use one focused modification at a time. Explicitly protect the parts that already work: "Keep the current camera movement and pacing unchanged. Change only the wardrobe color."
Plan a small number of targeted iterations. Each rerun or revision can be treated as a new generation request and consume additional credits, so refine your inputs before resubmitting.
Prompt Templates for Each Preset
Prompt structure decides whether a template performs well. Separate preserved elements from changed elements in every prompt. Adapt these templates to your project:
| Scenario | Prompt Template |
|---|---|
| First generation | Use the reference video as the foundation. Replace the main subject with [subject]. Preserve the original movement, camera direction, and pacing. Set the scene in [environment] with [style]. |
| Product replacement | Replace the product with [description]. Keep the original shot sequence, camera movement, and timing. Make the product clearly visible with [details]. Use a polished advertising presentation. |
| Character & wardrobe | Replace the character with [description]. Preserve the reference performance and body movement. Dress the character in [wardrobe], including [colors, materials, accessories]. |
| Style transfer | Apply a [style] to the reference video. Preserve composition, movement, and pacing. Change the palette to [colors], lighting to [description], atmosphere to [mood]. |
| Follow-up edit | Keep the current camera movement, pacing, and main action unchanged. Change only [target element]. Preserve [elements that already work]. |
Describe characters with visible traits, not vague labels. List wardrobe details in order of visual importance, and never combine two contradictory style directions in a single request.
Input Checklist Before Generating
Run through this checklist every time before you commit credits to a generation. It prevents the most common upload and consistency failures.
Pre-Generation Checklist:
- Reference video opens, plays, and shows the intended movement clearly
- Character, product, and clothing images are clean, consistent, and unobstructed
- Shared file links open without an extra login
- Video duration and image count fit the limits shown in the interface
- All uploaded media is owned or authorized for use and transformation
- Prompt separates preserved elements from changed elements
- Displayed credit estimate reviewed before submitting
Upload only media you own or are licensed to use—this applies to people, characters, brands, products, and music. For commercial work, keep records of required approvals.
FAQ
Q: What is the best Higgsfield Katana template for product ads?
Personalized Edits is the strongest starting point. Pair a reference video with a clear product image, keep the original shot sequence and timing, and describe the product's most important visible characteristics so it remains the visual focus.
Q: Can I combine Style Transfer and Motion Transfer in one request?
You can combine directions, but your prompt must clearly separate them. State that movement and pacing must be preserved while only the visual identity changes. Combining conflicting instructions in one request is the most common cause of results that drift from the reference.
Q: Do template reruns cost extra credits?
A rerun or revised generation is typically treated as a new request and may consume additional credits. Review the displayed credit estimate before each submission and refine your reference material and instructions before retrying.
Q: Why doesn't my generated video match the reference?
Reference transfer depends on action clarity, subject visibility, image consistency, and instruction specificity. Use a reference with one readable movement, supply consistent subject images, and iterate on one change at a time instead of rewriting the whole concept.