How to use and evaluate “Standing Still in the Crowd”
This street prompt is designed as an editable starting point, not a guaranteed one-click result. The notes below translate its JSON into a readable visual plan, identify the controls that matter, and show how to diagnose a weak generation without simply adding more text.
Prompt-specific blueprint
Scene and atmosphere
- Location
- a crowded urban street in Paris, France
- Time
- daytime
- Mood
- cinematic, energetic, fashionable, urban chaos
Visual treatment
- Photography
- ultra-realistic cinematic editorial street photography
- Camera
- camera positioned approximately 60 degrees above the subject, shallow depth of field, premium full-frame photography
- Lighting
- natural outdoor daylight with realistic contrast and soft ambient illumination
Composition and wardrobe
- People Count Label
- one uploaded reference person surrounded by a large moving crowd
- People Count
- multiple
- Reference Instruction
- Use one uploaded reference person as the main subject while all surrounding people are anonymous individuals with heavy motion blur.
- Wardrobe · Style
- modern oversized streetwear
- Wardrobe · Fit
- baggy oversized clothing
Recommended workflow
- 01
Prepare the input
Choose reference images you have permission to use. For Standing Still in the Crowd, prioritize a clear subject and avoid unrelated people or private details in the frame.
- 02
Lock the visual plan
Confirm subject count, framing, scene, and lighting before adding finishing language. Resolve conflicts such as close-up versus full-body composition.
- 03
Adapt for one model
Start with Midjourney. Keep a baseline generation and document the exact model version, references, aspect ratio, and visible settings.
- 04
Revise from evidence
Name the visible failure, change one instruction block, and compare it against the baseline. Keep only revisions that improve the intended criterion.
Model adaptation notes
Midjourney
Translate the core fields into concise natural language and add current version-specific reference or aspect controls in Midjourney itself.
Gemini
Provide the JSON with clearly labeled reference images. Restate any ignored constraint in plain language without duplicating the whole prompt.
ChatGPT Image
Keep the JSON structure, attach references in the same conversation, and use follow-up edits for one local correction at a time.
Stable Diffusion
Move observable failures into the interface’s negative prompt field and document the checkpoint, sampler, guidance, and resolution used.
Flux
Preserve the semantic hierarchy but favor coherent natural language; weighting and negative-prompt behavior depend on the host workflow.
A compatibility label means the visual intent can be adapted; it does not promise identical output across providers, versions, or settings.
Prompt-specific quality checklist
- subject remains perfectly sharp
- all surrounding pedestrians are heavily motion blurred
- strong sense of movement and urban chaos
- Paris street atmosphere
- direct eye contact with the camera
- realistic open skin pores
- subtle pigmentation
Failures to check before publishing
- plastic skin
- beauty filter
- over-retouching
- AI-looking face
- blurry main subject
- low-resolution skin
- duplicate people
Why the editable controls matter
Gender
Required identity direction for the reference subject.
People count
Default is a single uploaded reference person.
Identity or subject drift
Reduce competing style language, improve the reference, and separate stable traits from pose, gaze, and expression.
Composition breaks
Check whether crop, shot distance, aspect ratio, subject count, and required objects can all be satisfied in one frame.
Artificial lighting or texture
Name a motivated light source and believable material behavior before adding grain, grading, sharpness, or resolution terms.
@cankirkgoz
Comments
You need to sign in to comment.
No comments yet.