File size: 1,983 Bytes
a9e2910
e373289
 
 
a9e2910
 
 
e373289
2e7d4a3
a9e2910
 
e373289
 
 
 
 
 
 
 
 
 
 
 
c8c583c
 
 
e373289
 
 
 
 
 
 
c8c583c
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
---
title: smile-enhancement-ai-smg
emoji: 😁
colorFrom: indigo
colorTo: blue
sdk: gradio
app_file: app.py
pinned: true
sdk_version: 5.29.0
---

# Smile Enhancement AI

A simple, user-friendly application that uses Google's Gemini AI to enhance smiles in portrait photos. The app creates natural-looking smile enhancements with adjustable intensity levels.

## Features

- Simple upload interface - just drag and drop your photo
- Creates personalized, natural-looking smiles that suit the individual
- Preserves natural dental characteristics while making subtle enhancements
- Maintains each person's unique smile features and tooth coloration
- Enhances eye crinkles and cheek raising for authentic expressions
- Preserves facial identity while enhancing smiles
- Multi-attempt enhancement that compares results to select the best outcome
- Intelligent backend processing that may run multiple enhancement passes for better results
- Results may improve or degrade across multiple attempts as the system learns what works best
- Works with most portrait photos showing visible teeth
- No API key or login required

## Usage

1. Upload a clear selfie with a visible smile
2. Click "Enhance Smile with Natural Expressions" to process your image
3. Download your enhanced smile image

## How It Works

This application uses an advanced process that may run multiple enhancement attempts to achieve the best results:

- The first enhancement attempt might not always be the best result
- Our system intelligently runs multiple passes on your image (up to 3 attempts)
- Each attempt is assessed and compared to find the highest quality enhancement
- The best result is automatically selected and presented to you

Please note that results may vary as we haven't provided the system with extensive example data. Sometimes later attempts produce better results, while in other cases earlier attempts may be superior. The system intelligently selects the best outcome from all attempts.