AI AI Tools Guidebook
AI Tool Tutorials

AI Fantasy Character Design Tutorial: From Sheet to Splash

Build a fantasy character bible: turnaround sheet, environment splash, and action variant on one face.

Published: Author: AI Productivity Guide Team 🌐 查看中文版本

Fantasy character design done in a single prompt gives you a cool image and zero usability. What ships in a deck, a TTRPG sourcebook, or a game pitch is a three-asset bundle: a character sheet (turnaround + traits), an environment splash that proves the character can carry a scene, and an action variant that shows how they move. This tutorial builds all three on the same face, using Midjourney for ideation and SDXL with a character LoRA for downstream consistency.

What this covers

A three-asset fantasy character pipeline: turnaround sheet first, then splash, then action variant. The discipline is one canonical face, one frozen trait list, and one signature silhouette element that survives every scene.

Who this is for

TTRPG game-masters and indie publishers, narrative game devs prototyping cast, fantasy authors building a series bible, illustrators using AI for ideation before final rendering, and pitch-deck builders for animated shorts or pilots.

When to reach for it

A new TTRPG campaign needing 4-8 characters, a game pitch needing one hero across 3-5 hero shots, a series bible for a novel or webcomic, a Kickstarter campaign that needs visual cohesion across rewards, or any time you need one character to read as the same person across very different stagings.

Before you start

  • Decide rendering style early — painterly illustration, anime, semi-realist, dark-fantasy oil. Style choice constrains tooling and trait wording.
  • Sketch a one-paragraph character brief: name, role, background, signature silhouette element (cloak, weapon, mask, glow). The silhouette is your second face.
  • Choose toolchain: Midjourney --cref for fast ideation, SDXL + IP-Adapter or a 20-image-trained LoRA for tight consistency, Flux Redux for high-fidelity reference.
  • Reserve 2-3 hours for the canonical pass. The sheet is the most-reused asset; do not rush it.

Step by step

  1. Generate the character sheet first: front, three-quarter, side, and back views in one image, neutral pose, even lighting, gray studio backdrop. Use prompt: “character sheet, four views, T-pose, neutral expression, studio gray”.
  2. From the sheet, extract the trait list (5-7 visible items): face, hair, eyes, skin, signature outfit/armor, weapon, distinguishing mark. Paste verbatim into every downstream prompt.
  3. Identify the signature silhouette element — the one shape readable in a black-on-white silhouette. Could be a hooded cloak, a curved sword across the back, a glowing third eye. Keep this constant.
  4. Generate the environment splash. Same trait block, add scene: “[trait block] standing on a cliff overlooking a ruined city at dusk, dramatic side-lighting, painterly”. Use the character sheet as --cref reference.
  5. Generate the action variant. Same trait block, add action and dynamic camera: “[trait block] mid-leap with sword drawn, dynamic low angle, motion blur on cloak, dust kicked up”. Keep face anchored via reference image.
  6. Compare all three side by side. Squint test: same person? If the action variant face drifts, increase reference weight or regenerate.
  7. Save the canonical sheet + splash + action variant as the character bible v1. Future scenes reference this set.

First-run exercise

  1. Pick one character you actually need (a campaign NPC, a novel protagonist) — not a roster of six.
  2. Generate the four-view sheet first. Spend 60-90 minutes; the rest of the bible rides on this.
  3. Generate one splash and one action variant using the sheet as reference. Squint at all three at thumbnail size.
  4. If anything reads as a different character, name the failure (jaw shifted, cloak color drifted, weapon changed) and re-run that one variant.

Quality check

  • Same face across sheet, splash, and action variant at thumbnail size?
  • Is the signature silhouette element present and readable in all three (especially the action variant)?
  • Does the rendering style hold across scenes? Mixing painterly and photoreal mid-bible breaks the read.
  • Do distinguishing marks (scars, glyphs, jewelry) carry through? AI loves to drop these silently.
  • Does the action variant still read as in-style, or did dynamic posing nudge the model toward anime defaults?

How to reuse this workflow

  • Save the trait block + signature silhouette element as a character-bible.md. New scenes only add staging.
  • For a roster of 4-8 characters, run the sheet step for all of them first, then splashes, then action variants. Batching keeps the model “warmed up” per character.
  • Once you have 20+ approved scenes for a character, train a LoRA. Future generations stay consistent without needing the reference attached.
  • Re-test reference weight every 6 weeks; major model updates can require retuning.

Character brief paragraph → four-view sheet → trait extraction → environment splash with reference lock → action variant with reference lock → bible v1 saved → future scenes reference the bible. Total time for one character: 3-4 hours including iteration. Saves dozens of hours of “this doesn’t look like the same person” later.

Common mistakes

  • Skipping the sheet and going straight to a splash. Without a clean canonical, every downstream image drifts.
  • Letting the signature silhouette element wander. The cloak that defined the silhouette in scene 1 is missing by scene 4.
  • Rephrasing the trait block per scene. Lock the wording byte-identical — small synonym swaps compound.
  • Mixing rendering styles across bible assets. Painterly sheet + photoreal splash = two characters.
  • Dynamic action prompts that overwhelm the reference. Push reference weight up when the action gets extreme.
  • Building a roster of 8 characters in one session without saving canonicals. You lose the anchors and everything drifts.

FAQ

  • Sheet view count — four or eight?: Four (front, 3/4, side, back) is the sweet spot. Eight-view sheets confuse most models; they smear faces across the row.
  • Best tool for fantasy consistency?: Midjourney --cref for ideation, SDXL + custom LoRA for production. Flux Redux is strong for high-fidelity face matching across one or two scenes.
  • Can I generate the character holding different weapons?: Yes, but treat the weapon as a trait — if it changes, the silhouette read changes. Document weapon variants in the bible.
  • Should I train a LoRA?: Worth it past 20-30 generations for one character. Below that, reference image is faster.
  • Why does the face drift in the action variant?: Dynamic poses push the model toward defaults. Bump reference weight and consider generating the action variant as image-to-image from a posed sheet pose.

Tags: #Midjourney #sdxl #Fantasy #character-design #Tutorial

Related Articles