Spaces:
Sleeping
Sleeping
Update README.md
Browse files
README.md
CHANGED
@@ -1,124 +1,132 @@
|
|
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 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
|
54 |
-
|
55 |
-
|
56 |
-
|
57 |
-
|
58 |
-
|
59 |
-
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
69 |
-
|
70 |
-
|
71 |
-
|
72 |
-
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
|
83 |
-
|
84 |
-
|
85 |
-
"
|
86 |
-
|
87 |
-
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
-
|
98 |
-
|
99 |
-
|
100 |
-
|
101 |
-
|
102 |
-
|
103 |
-
|
104 |
-
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
-
|
114 |
-
|
115 |
-
|
116 |
-
|
117 |
-
|
118 |
-
|
119 |
-
|
120 |
-
|
121 |
-
|
122 |
-
|
123 |
-
|
124 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
title: FastAPI Model
|
3 |
+
sdk: docker
|
4 |
+
emoji: π
|
5 |
+
colorFrom: blue
|
6 |
+
colorTo: purple
|
7 |
+
pinned: false
|
8 |
+
---
|
9 |
+
<<<<<<< HEAD
|
10 |
+
# AI-Powered Body Measurement & Apparel Sizing
|
11 |
+
|
12 |
+
This project is a **FastAPI-based AI system** that predicts **body measurements** and recommends **clothing sizes** using deep learning. It processes **front and side images**, removes backgrounds, and utilizes a trained **Keras model** for accurate predictions.
|
13 |
+
|
14 |
+
---
|
15 |
+
|
16 |
+
## π Features
|
17 |
+
β
AI-powered **body measurement estimation** using images.
|
18 |
+
β
Automatic **background removal** for clean image processing.
|
19 |
+
β
Deep learning-based **predictions** using a trained model.
|
20 |
+
β
**FastAPI integration** for quick and efficient API responses.
|
21 |
+
|
22 |
+
---
|
23 |
+
|
24 |
+
|
25 |
+
## π Installation
|
26 |
+
|
27 |
+
To set up the project, follow these steps:
|
28 |
+
|
29 |
+
### **1οΈβ£ Clone the Repository**
|
30 |
+
```bash
|
31 |
+
git clone https://github.com/CloozyBrands/AI-BodyMeasurement.git
|
32 |
+
```
|
33 |
+
---
|
34 |
+
### **2οΈβ£ Install Dependencies**
|
35 |
+
```bash
|
36 |
+
pip install -r requirements.txt
|
37 |
+
```
|
38 |
+
|
39 |
+
### **3οΈβ£ Run the FastAPI Server**
|
40 |
+
```bash
|
41 |
+
uvicorn app.main:app --reload
|
42 |
+
```
|
43 |
+
Then open **http://127.0.0.1:8000/docs** to test the API.
|
44 |
+
|
45 |
+
---
|
46 |
+
|
47 |
+
## π οΈ How It Works
|
48 |
+
|
49 |
+
### **1οΈβ£ Upload Images**
|
50 |
+
- The API accepts **two images**:
|
51 |
+
- **Front view**
|
52 |
+
- **Side view**
|
53 |
+
|
54 |
+
### **2οΈβ£ AI-Based Processing**
|
55 |
+
- The images are processed using **`single_person_processor.py`**, which:
|
56 |
+
β
Removes the background using `rembg`.
|
57 |
+
β
Converts the images into a **model-compatible format**.
|
58 |
+
β
Predicts **body measurements** like chest, waist, and height.
|
59 |
+
|
60 |
+
### **3οΈβ£ Clothing Size Prediction**
|
61 |
+
- Based on the body measurements, the system suggests a **T-shirt and pants size** using predefined **size charts**.
|
62 |
+
|
63 |
+
---
|
64 |
+
|
65 |
+
## π API Endpoints
|
66 |
+
|
67 |
+
### **1οΈβ£ `/predict/` (POST) - Predict Body Measurements**
|
68 |
+
π **Example Request:**
|
69 |
+
```http
|
70 |
+
POST /predict/
|
71 |
+
```
|
72 |
+
|
73 |
+
π **Request Parameters:**
|
74 |
+
| Parameter | Type | Description |
|
75 |
+
|---------------|--------|-------------|
|
76 |
+
| `front_image` | File | Front view image (JPEG/PNG) |
|
77 |
+
| `side_image` | File | Side view image (JPEG/PNG) |
|
78 |
+
| `input_data` | JSON | User data (height, weight, gender) |
|
79 |
+
|
80 |
+
π **Example JSON Payload:**
|
81 |
+
```json
|
82 |
+
{
|
83 |
+
"gender": 0,
|
84 |
+
"height_cm": 175,
|
85 |
+
"weight_kg": 70,
|
86 |
+
"apparel_type": "all"
|
87 |
+
}
|
88 |
+
```
|
89 |
+
|
90 |
+
π **Example Response:**
|
91 |
+
```json
|
92 |
+
{
|
93 |
+
"results": {
|
94 |
+
"body_measurements": {
|
95 |
+
"chest": 100.5,
|
96 |
+
"waist": 80.2,
|
97 |
+
"hip": 97.3
|
98 |
+
},
|
99 |
+
"tshirt_size": "L",
|
100 |
+
"pants_size": 34
|
101 |
+
}
|
102 |
+
}
|
103 |
+
```
|
104 |
+
|
105 |
+
---
|
106 |
+
|
107 |
+
## π― Model & AI Processing
|
108 |
+
|
109 |
+
π The model used in this project is a **TensorFlow/Keras** model stored as `best_model.keras`.
|
110 |
+
π The AI processing is handled inside **`single_person_processor.py`**, which:
|
111 |
+
- Loads the trained model using `tf.keras.models.load_model`.
|
112 |
+
- Extracts measurements based on input images.
|
113 |
+
- Maps the measurements to standard clothing sizes.
|
114 |
+
|
115 |
+
---
|
116 |
+
|
117 |
+
## π License
|
118 |
+
|
119 |
+
This project is licensed under the **MIT License** β feel free to modify and use it.
|
120 |
+
|
121 |
+
---
|
122 |
+
|
123 |
+
## π€ Contributing
|
124 |
+
|
125 |
+
1. Fork the repository
|
126 |
+
2. Create a new branch
|
127 |
+
3. Make your changes
|
128 |
+
4. Submit a pull request
|
129 |
+
|
130 |
+
We welcome contributions and improvements! π
|
131 |
+
```
|
132 |
+
>>>>>>> 4e80050 (FastAPI model for size recommendation)
|