Crafting Realistic Smartphone Selfie Textures: A Structured JSON Prompt Engineering Guide

On October 5, 2026, visual creator Viggo Hayes (@ViggoHayes) shared a structured JSON prompt template engineered to recreate the authentic skin texture, lens perspective, and candid digital noise of a casual smartphone front-camera selfie. By systematically eliminating artificial CG plastic sheen and sterile studio lighting, the workflow demonstrates how object-oriented prompt hierarchies can reliably produce intimate, alternative-fashion social media portraits with believable photographic fidelity.

Image source: X @ViggoHayes

When generating character portraits with traditional natural-language prompts, modern image synthesis models often default to over-polished commercial beauty photography or glossy 3D rendering. This structured approach replaces long descriptive sentences with an explicit key-value JSON schema. By compartmentalizing the subject, camera optics, background props, and negative constraints into dedicated modules, the prompt prevents semantic interference and guides the model toward authentic digital photography aesthetics.

Complete Structured JSON Prompt Template

The complete verbatim JSON prompt shared by the creator is presented below. It partitions all visual dimensions into specialized configuration blocks for deterministic control.

{
  "subject": {
    "description": "A young adult woman taking an extremely close handheld selfie indoors. She has a short silver-lavender bob haircut with straight blunt bangs, large dark eyes, and a soft pale complexion. She wears an all-black gothic-inspired outfit with layered ruffles and lace. Her face dominates the frame, creating an intimate, slightly mischievous alternative-fashion portrait.",
    "mirror_rules": {
      "type": "front-facing handheld selfie",
      "device_position": "camera held very close to the face, slightly above eye level",
      "reflection_logic": "direct camera selfie rather than a mirror reflection"
    },
    "age": "Young adult visual impression; preserve the youthful styling without making a specific age claim",
    "expression": {
      "eyes": {
        "look": "directly into the camera",
        "energy": "quiet, playful, slightly mischievous",
        "direction": "straight toward the lens"
      },
      "mouth": {
        "position": "slightly open with a playful expression",
        "energy": "cheeky and teasing rather than overtly sensual"
      },
      "overall": "A playful alternative-fashion selfie with an intentionally provocative-looking but non-explicit facial expression"
    },
    "face": {
      "preserve_original": true,
      "makeup": "minimal Japanese alternative-fashion makeup, softly defined eyes, subtle lower-eye shading, very light blush, natural pale lips, luminous but realistic skin"
    }
  },
  "hair": {
    "color": "very pale silver-lavender, almost white with a cool lilac tint",
    "style": "short rounded bob with heavy straight bangs reaching toward the eyebrows and partially framing both sides of the face",
    "effect": "smooth, fine, slightly wispy synthetic-dyed texture with soft cool highlights and a few naturally separated strands"
  },
  "body": {
    "frame": "slim upper-body portrait, with the shoulders and torso filling the lower frame",
    "waist": "not visible",
    "chest": "mostly covered by layered black clothing",
    "legs": "not visible",
    "skin": {
      "visible_areas": [
        "face",
        "neck",
        "shoulders",
        "upper chest around the neckline",
        "hands and fingers"
      ],
      "tone": "very fair neutral-cool complexion",
      "texture": "soft youthful skin with subtle natural pores, faint tonal variation, and realistic facial texture",
      "lighting_effect": "soft diffuse indoor light creates gentle highlights on the forehead, nose, cheeks, and fingers without strong directional shadows"
    }
  },
  "pose": {
    "position": "upper body close to the camera",
    "base": "one hand is raised toward the lower face, with the fingers positioned near the lips in a playful gesture",
    "overall": "close, slightly hunched selfie posture with the head tilted subtly toward the camera and shoulders relaxed"
  },
  "clothing": {
    "top": {
      "type": "black gothic-inspired layered blouse or dress",
      "color": "deep matte black",
      "details": "ruffled shoulders, gathered fabric, lace panels, layered neckline, loose sleeves, subtle straps and textured black-on-black detailing",
      "effect": "soft matte fabric with deep folds and rich shadow separation between the lace and outer layers"
    },
    "bottom": {
      "type": "not visible",
      "color": "not visible",
      "details": "not applicable"
    }
  },
  "accessories": {
    "headwear": "none clearly visible",
    "jewelry": "minimal dark necklace or thin accessory partially visible near the neckline",
    "device": "front-facing smartphone camera, device outside the visible frame",
    "prop": "none"
  },
  "photography": {
    "camera_style": "casual front-camera smartphone selfie with an early social-media / Japanese alternative-fashion photo aesthetic",
    "angle": "slightly high angle, very close to the face",
    "shot_type": "extreme close-up portrait / tight upper-body selfie",
    "aspect_ratio": "approximately 4:5 vertical",
    "texture": "slightly soft digital image, mild compression, subtle sensor noise, gentle highlight bloom, natural smartphone processing",
    "lighting": "soft overhead indoor ambient light with weak frontal fill, relatively flat illumination and gentle shadows under the chin and clothing folds",
    "depth_of_field": "moderate depth of field; face and foreground remain sharp while the room behind her is somewhat softer"
  },
  "background": {
    "setting": "casual indoor dressing room, bedroom, backstage room, or alternative-fashion studio",
    "wall_color": "off-white and pale gray",
    "elements": [
      "large mirror",
      "blurred person reflected in the mirror",
      "dark clothing",
      "black boots",
      "guitar or musical equipment",
      "casual room objects"
    ],
    "atmosphere": "unpolished, intimate, candid, slightly underground alternative-fashion atmosphere",
    "lighting": "ordinary cool-neutral indoor lighting without dramatic studio illumination"
  },
  "the_vibe": {
    "energy": "playful, mischievous, intimate",
    "mood": "quietly rebellious and cute with a slightly strange internet-selfie quality",
    "aesthetic": "Japanese alternative fashion, gothic lolita-adjacent styling, silver-lavender hair, black lace, nostalgic social-media selfie",
    "authenticity": "retain imperfect framing, ordinary indoor lighting, mild compression, slight softness, visible background clutter, and natural hair irregularities",
    "intimacy": "very close camera distance makes the portrait feel personal and spontaneous",
    "story": "A quick selfie taken indoors before or after an alternative-fashion event, with the subject deliberately making a playful expression for the camera.",
    "caption_energy": "just being a little strange today"
  },
  "constraints": {
    "must_keep": [
      "very pale silver-lavender bob",
      "heavy straight bangs",
      "large dark eyes",
      "very close selfie composition",
      "black gothic-inspired clothing",
      "ruffled and lace textures",
      "fair cool-toned skin",
      "direct eye contact",
      "playful hand-near-mouth gesture",
      "ordinary indoor background",
      "large mirror",
      "casual smartphone image quality",
      "slightly imperfect framing",
      "soft diffuse lighting",
      "alternative Japanese fashion atmosphere"
    ],
    "avoid": [
      "commercial studio portrait",
      "glamour photography",
      "perfect beauty retouching",
      "plastic skin",
      "CG hair",
      "excessive makeup",
      "overly sexualized styling",
      "dramatic cinematic lighting",
      "clean empty background",
      "perfect symmetry",
      "extra fingers",
      "deformed hands",
      "extra arms",
      "incorrect hair color",
      "bright colorful clothing"
    ]
  },
  "negative_prompt": [
    "studio portrait",
    "glamour shoot",
    "plastic skin",
    "CG hair",
    "heavy makeup",
    "oversexualized",
    "dramatic lighting",
    "clean background",
    "perfect symmetry",
    "extra fingers",
    "deformed hands",
    "extra arms",
    "wrong hair color",
    "bright clothing",
    "excessive retouching"
  ]
}

Modular Analysis and Realism Engineering Techniques

The effectiveness of this prompt template stems from its granular modeling of smartphone camera optics, facial physics, and ambient surroundings.

1. Camera Optics and Artifact Injection (photography)

• Focal Distance and Perspective: Defining an extreme close-up (held very close to the face, slightly above eye level) forces the subject's gaze and facial expression to dominate the frame, replicating the natural slight wide-angle expansion typical of smartphone front cameras.
• Sensor Noise and Digital Processing: Rather than prompting for ultra-sharp 8K fidelity, the schema intentionally requests a slightly soft digital image, mild compression, subtle sensor noise, and gentle highlight bloom. These parameters mimic the built-in computational photography tone mapping of early-to-mid smartphone generations.
• Disambiguating Selfies from Mirror Shots: The explicit reflection_logic rule (direct camera selfie rather than a mirror reflection) prevents the model from generating unwanted phone hardware in front of the face while still allowing background mirrors to show natural reflections.

2. Lifelike Skin Physics and Facial Expressions (skin, expression)

• Preserving Micro-Textures: To defeat airbrushed porcelain skin, the prompt specifies subtle natural pores, faint tonal variation, and realistic facial texture. This creates authentic specular response under indoor lighting.
• Understated Subculture Makeup: Aligning with Japanese alternative fashion aesthetics, the makeup instructions focus on soft eye contouring, delicate under-eye shading, and pale natural lips, preventing the heavy, cakey cosmetic styling often produced by beauty presets.
• Dynamic Hand Interaction: Anchoring the hand near the lips (hand is raised toward the lower face, with the fingers positioned near the lips) introduces natural spatial depth and human interaction into an otherwise static portrait.

3. Candid Environmental Context (background, the_vibe)

• Authentic Cluttered Backdrops: By specifying a dressing room or backstage space with a large mirror, casual dark clothing, boots, and musical equipment, the prompt grounds the portrait in a credible physical environment rather than an empty void.
• Accepting Imperfections: The template explicitly embraces imperfect framing, visible background clutter, and ordinary overhead fluorescent ambient illumination (soft overhead indoor ambient light), stripping away the artificial perfection that typically flags an image as AI-generated.

Defensive Guardrails via Explicit Constraints

The dual protection of constraints.avoid and negative_prompt acts as a deterministic barrier against recurring generative flaws:

• Banning Commercial Stereotypes: Explicitly excluding glamour shoots, cinematic lighting, and studio backdrops steers the diffusion model away from glossy advertising tropes.
• Suppressing Plastic Artifacts: Blacklisting plastic skin and CG hair directly targets synthetic shader artifacts in the model's latent weights.
• Anatomy Verification: Pinpointing extra fingers and deformed hands provides an essential check when generating complex poses where fingers closely interact with facial features.

By structuring visual parameters into an organized JSON schema, creators can enforce precise aesthetic nuances and achieve believable social-media snapshot realism without fighting unpredictable prompt degradation.

Original source

• Original post on X (Twitter): @ViggoHayes Post (2107049918137598433)