DECODE · AI IMAGE ANALYSIS

Decode

Upload any image and AIKIZI breaks it down across 7 dimensions: mood, composition, style influences, materials, and more. You get ready-to-use AI prompts, copyable style keywords, and deep insights in about 15 seconds.

FLOW

How it works

01
UPLOAD

Upload

Drop any image into the Studio. JPG, PNG, or WebP up to 50MB.

02
DECODE

Decode

AI analyzes your image across 7 dimensions in about 15 seconds.

03
USE

Explore and use

Browse tabs, copy prompts, grab style keywords, generate new images.

DIMENSIONS

The 7 analysis tabs

Each tab covers a different dimension of the image. Together they form the complete breakdown.

CORE
The essentials: subject, scene, titles, and ready-to-use captions. Includes three evocative titles, the three-word Style Triplet, a scene synopsis, activity, captions, and professional use cases.
MOOD
The emotional read: mood, tension, and psychological depth. Three ranked mood descriptors, emotional resonance with tension points, inner state, and dominant expression.
VISUAL
Camera and light: angles, lighting setup, and the complete color palette. Frame, vertical angle, light source and quality, plus dominant colors with copyable hex codes and a color signature.
FORM
Body and detail: pose, expression, hair, makeup, garments, and accessories. Subject type classification, character profile, pose with anatomical landmarks, and worn items.
STYLE
Artistic DNA: the visual fingerprint. Seven two-word Core Tokens for AI generators, layered Style DNA, aesthetic tags, era tags, historical roots, and a style lexicon.
WORLD
Context and culture: where and when. Cultural context, geographic region markers, era indicators, and mythological or symbolic motifs.
MATTER
Materials and craft: what things are made of. Material types, surface finish, micro-texture analysis, construction methods, and craftsmanship details.
OUTPUT

What you can do with a decode

Copy style keywords

Grab the 7 Core Tokens from STYLE and paste them into any AI generator.

Generate new images

Use the CREATE prompt directly in AIKIZI or copy it into your own tool.

Unlock SREF codes

Any account unlocks MidJourney style reference codes (5 credits each) for exact aesthetics.

Publish to Explore

Share your decode publicly so others can discover the breakdown.

Download analysis

Export the full analysis as JSON or copy individual sections.

Build on the palette

Copy hex codes from VISUAL to use exact colors in your designs.

Every decode also generates a ready-to-use AI generation prompt crafted from all the analysis data. Copy it into MidJourney, DALL-E, Stable Diffusion, or use it in AIKIZI's built-in generator. You also get story ideas, dialogue suggestions, music pairings, and style remix ideas.

CONCEPTS

Key concepts

CORE TAB

Style Triplet

A three-word label capturing visual style essence, such as "Baroque Chiaroscuro Elegance" or "Dutch Golden Warmth". You get three in the CORE tab, great for searching, comparing, and describing styles at a glance.

STYLE TAB

Core Tokens

Seven two-word phrases capturing visual DNA, like "dramatic lighting", "golden tones", or "deep shadows". Copy them directly into AI generators for consistent style reproduction.

STYLE TAB

Style DNA

A recipe for the image's look, broken into layers: Foundation (40-60%), Primary (20-30%), Secondary (10-20%), and Accent (5-15%). It runs from dominant foundation to subtle accents.

5 CREDITS

SREF Codes

Some decodes include a locked MidJourney style reference code that recreates the exact visual style. On any account, open the decode, find the SREF section, tap Unlock, and copy the code. Once unlocked, it stays available permanently.

PRO

Video and audio decode

Pro subscribers can also analyze videos and audio files. The tabs adapt: VISUAL becomes CINEMATIC, FORM becomes KINETIC, and MATTER becomes PHYSICS. Audio has its own set: SONIC, HARMONIC, TEXTURE, and more.

IMAGE
5 credits. Available to all users.
VIDEO
15 credits. Pro only.
AUDIO
10 credits. Pro only.

If you are holding an image and want its style named, start with What art style is this?, three published decodes read in four passes. Related guides: Realms for AI generation templates.