Photo Guidelines

The quality of the input image directly affects the quality of the processed output. Follow these guidelines to ensure optimal results from the SmileGeni 2D API.

Supported Formats

PNGJPGJPEGGIFBMPWEBP
Recommended Format

PNG and JPG are recommended for best compatibility and processing speed. Use high-quality compression settings when saving JPG files.

Image Composition

The following composition requirements help the AI accurately detect and process facial features.

Face Position

The subject's face should be centered in the frame and take up a significant portion of the image. Avoid extreme cropping at the chin or forehead.

Angle

Front-facing photos produce the best results. The subject should look directly at the camera. Slight angles are acceptable, but avoid profile or extreme side views.

Teeth Visibility

The mouth and teeth should be clearly visible. A natural smile or slightly open mouth works best. The teeth area must not be obscured by objects, hands, or accessories.

Lighting

Use even, diffused lighting. Natural daylight or soft indoor lighting produces the best results. Avoid harsh shadows, backlighting, or overly bright spots on the face.

Resolution

Higher resolution images produce better results. A minimum of 512x512 pixels is recommended. Modern smartphone photos typically exceed this requirement.

Background

Background complexity does not significantly affect processing. The AI focuses on facial regions. However, a clean background helps with overall image quality.

Do's and Don'ts

Do

  • Use a front-facing photo with the subject looking directly at the camera
  • Ensure the mouth and teeth area is clearly visible
  • Use even, natural lighting on the face
  • Keep the face centered in the frame
  • Use a high-resolution image for the best results
  • Capture with a neutral or slightly smiling expression
  • Use photos taken with a smartphone or professional camera

Don't

  • Use images with heavy shadows across the face
  • Submit photos where the mouth is obscured (hands, masks, objects)
  • Use extremely low-resolution or heavily compressed images
  • Include multiple faces in a single image without cropping
  • Use photos with extreme angles (looking far up, down, or to the side)
  • Submit images with heavy filters or extreme post-processing
  • Use photos with strong backlighting that silhouettes the face

AI-Strength Parameter

The AI-Strength parameter (range: 1–10, default: 3) controls the number of processing iterations applied to the image.

  • Low values (1–3): Subtle, natural-looking enhancement. Recommended for high-quality input images.
  • Medium values (4–6): Moderate enhancement. Good for average-quality input images.
  • High values (7–10): Aggressive processing. Useful for lower-quality inputs but may produce less natural results.
Start Low

Begin with the default value of 3 and increase incrementally if the results need more processing. Higher values are not always better — they depend on the input image quality.

Whitening Parameters

When using the whitening workflow, three parameters allow fine-grained control:

  • whitening_strength (0.0–1.0): Overall whitening intensity. Higher values produce whiter results.
  • whitening_brightness_boost (1.0–2.0): Multiplier for teeth brightness. Values above 1.5 may appear unnatural on some images.
  • whitening_yellow_reduction (0.0–1.0): Reduces yellow tones in the teeth area. Higher values remove more yellow.
Natural-Looking Results

For the most natural-looking whitening, use moderate values: whitening_strength=0.7, whitening_brightness_boost=1.3, and whitening_yellow_reduction=0.5.