FIAT 0.6 업데이트

README.md

@@ -1 +1,114 @@
-# FIAT-Release
+
+<img src="https://github.com/gellston/FIAT/blob/main/snapshoot/icons8_price_tag_96px.png?raw=true" width=40 height=40></img>
+FIAT 0.6 (Fast Image Annotation Tool)
+=======================
+
+FIAT is a free image labeling tool developed in C# WPF based on Visual Studio 2022 <br/>
+The current version supports labeling classification datasets and provides a PyTorch dataset loader.
+
+Development Environment
+=======================
+ - **Visual Studio 2022**
+ - **Microsoft .NET 6**
+
+
+Download
+=======================
+
+- <a href="https://github.com/gellston/FIAT/releases/download/0.6/FIAT.exe" target="_blank">FIAT 0.6 download</a>
+- <a href="https://dotnet.microsoft.com/en-us/download/dotnet/thank-you/runtime-desktop-6.0.2-windows-x64-installer" target="_blank">.NET6 framework</a>
+
+Reference
+=======================
+- <a href="https://github.com/gellston/FIAT/tree/main/example_dataset" target="_blank">FIAT dataset example</a>
+- <a href="https://github.com/gellston/FIAT/blob/main/python/torch_classification_FIATC_test.py" target="_blank">PyTorch training loop example (classification)</a>
+
+DEMO
+=======================
+- Classification
+<center>
+<img src="https://github.com/gellston/FIAT/blob/0.6/snapshoot/FIAT-example.gif?raw=true"></img>
+</center>
+
+Menu
+=======================
+<center>
+<img src="https://github.com/gellston/FIAT/blob/0.6/snapshoot/how%20to%20use%20and%20shortcut%20key.jpg?raw=true"></img>
+</center>
+
+Shortcut Key
+=======================
+* F1~F12 : Label the image with the target label at that index
+* Ctrl + S : Save all label information
+* Ctrl + O : Open image folder
+* Up : Previous image
+* Down : Next image
+
+PyTorch training loop example
+=======================
+``` python
+import torch
+import torch.nn as nn
+
+from torch.utils.data import DataLoader
+
+from util.FIATClassificationDataset import FIATClassificationDataset
+
+
+batch_size = 1
+training_epochs = 1
+
+datasets = FIATClassificationDataset('C://Github//FIAT//example_dataset//food//',   #FIAT dataset path
+                                     label_height=224,  #image height
+                                     label_width=224,   #image width
+                                     isColor=True,      #color load flag
+                                     isNorm=False)      #0~1 normalization flag
+
+data_loader = DataLoader(datasets, batch_size=batch_size, shuffle=True)
+
+for epoch in range(training_epochs):
+    avg_cost = 0
+    avg_acc = 0
+    total_batch = len(data_loader)
+
+    for x_input, y_input in data_loader:
+        print('x_input shape = ', x_input.shape)
+        print('y_input shape = ', y_input.shape)
+
+print('Training loop finished')
+
+```
+``` console
+C:\Python\python.exe C:/Github/FIAT/python/torch_classification_FIATC_test.py
+{'Color': '#FFFF0000', 'Name': 'Bread'}
+{'Color': '#FF008000', 'Name': 'Pizza'}
+{'Color': '#FFFFFFFF', 'Name': 'Hamburger'}
+{'Color': '#FFF79646', 'Name': 'Chicken'}
+x_input shape =  torch.Size([1, 3, 224, 224])
+y_input shape =  torch.Size([1, 4])
+x_input shape =  torch.Size([1, 3, 224, 224])
+y_input shape =  torch.Size([1, 4])
+```
+
+
+Strcture
+=======================
+### Classification
+<img src="https://github.com/gellston/FIAT/blob/main/snapshoot/snapshot1.jpg?raw=true"></img>
+
+- __target_info.json 
+    - File containing representative label information
+    - <img src="https://github.com/gellston/FIAT/blob/main/snapshoot/__target_info.jpg?raw=true" width=300></img>
+
+- (each file).json 
+    - A file containing user-labeled information about an image.
+    - <img src="https://github.com/gellston/FIAT/blob/main/snapshoot/each_file.jpg?raw=true" width=600></img>
+
+
+
+
+<div style="text-align: right; margin-right:30px;"> 
+
+[TOP](#vision-studio) 
+
+</div>

example_dataset/food/1_1_bread.json

@@ -0,0 +1 @@
+{"FileName":"1_1_bread.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\1_1_bread.jpg","ClassCollection":[{"Color":"#FFFF0000","Name":"Bread"}]}

example_dataset/food/1_1_friedchicken.json

@@ -0,0 +1 @@
+{"FileName":"1_1_friedchicken.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\1_1_friedchicken.jpg","ClassCollection":[]}

example_dataset/food/1_1_hamburger.json

@@ -0,0 +1 @@
+{"FileName":"1_1_hamburger.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\1_1_hamburger.jpg","ClassCollection":[]}

example_dataset/food/1_1_pizza.json

@@ -0,0 +1 @@
+{"FileName":"1_1_pizza.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\1_1_pizza.jpg","ClassCollection":[{"Color":"#FF008000","Name":"Pizza"}]}

example_dataset/food/2_1_bread.json

@@ -0,0 +1 @@
+{"FileName":"2_1_bread.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\2_1_bread.jpg","ClassCollection":[{"Color":"#FFFF0000","Name":"Bread"}]}

example_dataset/food/2_1_friedchicken.json

@@ -0,0 +1 @@
+{"FileName":"2_1_friedchicken.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\2_1_friedchicken.jpg","ClassCollection":[{"Color":"#FFF79646","Name":"Chicken"}]}

example_dataset/food/2_1_hamburger.json

@@ -0,0 +1 @@
+{"FileName":"2_1_hamburger.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\2_1_hamburger.jpg","ClassCollection":[{"Color":"#FFFFFFFF","Name":"Hamburger"}]}

example_dataset/food/2_1_pizza.json

@@ -0,0 +1 @@
+{"FileName":"2_1_pizza.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\2_1_pizza.jpg","ClassCollection":[{"Color":"#FF008000","Name":"Pizza"}]}

example_dataset/food/3_1_bread.json

@@ -0,0 +1 @@
+{"FileName":"3_1_bread.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\3_1_bread.jpg","ClassCollection":[{"Color":"#FFFF0000","Name":"Bread"}]}

example_dataset/food/3_1_friedchicken.json

@@ -0,0 +1 @@
+{"FileName":"3_1_friedchicken.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\3_1_friedchicken.jpg","ClassCollection":[{"Color":"#FFF79646","Name":"Chicken"}]}

example_dataset/food/3_1_pizza.json

@@ -0,0 +1 @@
+{"FileName":"3_1_pizza.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\3_1_pizza.jpg","ClassCollection":[]}

example_dataset/food/4_1_bread.json

@@ -0,0 +1 @@
+{"FileName":"4_1_bread.jpg","FilePath":"C:\\Users\\wantr\\Desktop\\food\\4_1_bread.jpg","ClassCollection":[]}

example_dataset/food/___target_info.json

@@ -0,0 +1 @@
+[{"Color":"#FFFF0000","Name":"Bread"},{"Color":"#FF008000","Name":"Pizza"},{"Color":"#FFFFFFFF","Name":"Hamburger"},{"Color":"#FFF79646","Name":"Chicken"}]