Upload 13 files

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+after_train/labels.jfif filter=lfs diff=lfs merge=lfs -text
+after_train/results.png filter=lfs diff=lfs merge=lfs -text
+result.gif filter=lfs diff=lfs merge=lfs -text
+result/result_2.mp4 filter=lfs diff=lfs merge=lfs -text
+result/result.mp4 filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,12 +1,29 @@
----
-title: Pokemon2
-emoji: 📈
-colorFrom: green
-colorTo: indigo
-sdk: gradio
-sdk_version: 5.49.1
-app_file: app.py
-pinned: false
----
+# Pokemon object detections
+Using yolov8 after training with Google Colab
+## Dataset
+Because of lack of data, there are only 7 classes: ```pikachu```, ```charmander```, ```bulbasaur```, ```squirtle```, ```eevee```, ```jigglypuff``` and ```other```.  
+  
+![labels](https://github.com/vovod/yolov8-pokemon-object-detection/blob/main/after_train/labels.jfif)
+## Requirements
+```
+pip install ultralytics
+```
+## Preprocess Data
+The ```convert.py``` used to convert *.xml* label file to *.txt* yolo label file.  
+Run ```resize_image.py``` to resize image's width to 640.
+## Train with Colab
+Edit ```name.yaml```.
+Upload images and labels.
+```
+!yolo train model=yolov8n.pt data=/content/name.yaml epochs=50 imgsz=640
+```
+## Training's Result
+```last.pt``` and ```best.pt``` in result folder.
+  
+![train](https://github.com/vovod/yolov8-pokemon-object-detection/blob/main/after_train/results.png?raw=true)
+## Predict
+Run ```predict.py``` to see result. This is my predict to ```test.mp4```:  
+  
+![result](https://github.com/vovod/yolov8-pokemon-object-detection/blob/main/result.gif?raw=true)
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+#### Thank you for stopping by!

after_train/labels.jfif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ab28834a9d80d80bf222ea9bd195ba3fb43fa28778fa478106a1bad4d69df5c
+size 233258

convert.py

@@ -0,0 +1,77 @@
+# -*- coding: utf-8 -*-
+
+from xml.dom import minidom
+import os
+import glob
+
+lut={}
+lut["pikachu"]=0
+lut["charmander"]=1
+lut["bulbasaur"]=2
+lut["squirtle"]=3
+lut["eevee"]=4
+lut["other"]=5
+lut["jigglypuff"]=6
+
+
+def convert_coordinates(size, box):
+    if(size[0]==0 or size[1]==0):
+        return (0,0,0,0)
+    dw = 1.0/size[0]
+    dh = 1.0/size[1]
+    x = (box[0]+box[1])/2.0
+    y = (box[2]+box[3])/2.0
+    w = box[1]-box[0]
+    h = box[3]-box[2]
+    x = x*dw
+    w = w*dw
+    y = y*dh
+    h = h*dh
+    return (x,y,w,h)
+
+
+def convert_xml2yolo(lut):
+
+    for fname in glob.glob("*.xml"):
+        
+        xmldoc = minidom.parse(fname)
+        
+        fname_out = (fname[:-4]+'.txt')
+
+        with open(fname_out, "w") as f:
+
+            itemlist = xmldoc.getElementsByTagName('object')
+            size = xmldoc.getElementsByTagName('size')[0]
+            width = int((size.getElementsByTagName('width')[0]).firstChild.data)
+            height = int((size.getElementsByTagName('height')[0]).firstChild.data)
+
+            for item in itemlist:
+                # get class label
+                classid =  (item.getElementsByTagName('name')[0]).firstChild.data
+                if classid in lut:
+                    label_str = str(lut[classid])
+                else:
+                    label_str = "-1"
+                    print ("warning: label '%s' not in look-up table" % classid)
+
+                # get bbox coordinates
+                xmin = ((item.getElementsByTagName('bndbox')[0]).getElementsByTagName('xmin')[0]).firstChild.data
+                ymin = ((item.getElementsByTagName('bndbox')[0]).getElementsByTagName('ymin')[0]).firstChild.data
+                xmax = ((item.getElementsByTagName('bndbox')[0]).getElementsByTagName('xmax')[0]).firstChild.data
+                ymax = ((item.getElementsByTagName('bndbox')[0]).getElementsByTagName('ymax')[0]).firstChild.data
+                b = (float(xmin), float(xmax), float(ymin), float(ymax))
+                bb = convert_coordinates((width,height), b)
+                #print(bb)
+
+                f.write(label_str + " " + " ".join([("%.6f" % a) for a in bb]) + '\n')
+
+        print ("wrote %s" % fname_out)
+
+
+
+def main():
+    convert_xml2yolo(lut)
+
+
+if __name__ == '__main__':
+    main()

name.yaml

@@ -0,0 +1,28 @@
+# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
+# COCO128 dataset https://www.kaggle.com/ultralytics/coco128 (first 128 images from COCO train2017) by Ultralytics
+# Example usage: python train.py --data coco128.yaml
+# parent
+# ├── yolov5
+# └── datasets
+#     └── coco128  ← downloads here (7 MB)
+
+
+# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
+path: ../datasets/name  # dataset root dir
+train: images/train  # train images (relative to 'path') 128 images
+val: images/train  # val images (relative to 'path') 128 images
+test:  # test images (optional)
+
+# Classes
+names:
+  0: pikachu
+  1: charmander
+  2: bulbasaur
+  3: squirtle
+  4: eevee
+  5: other
+  6: jigglypuff
+
+
+# Download script/URL (optional)
+download: https://ultralytics.com/assets/coco128.zip

predict.py

@@ -0,0 +1,11 @@
+import ultralytics
+from ultralytics import YOLO
+ultralytics.checks()
+import torch
+
+device = 'cuda'
+
+model = YOLO("best.pt")
+path ="test.mp4"
+
+results = model.predict(source=path, show = True)   

resize_image.py

@@ -0,0 +1,22 @@
+from PIL import Image
+import os
+from tqdm import tqdm
+
+in_file = "E:\\data\\pokemon_classify\\images\\test\\"
+out_file = "E:\\data\\pokemon_classify\\resize_image\\test\\"
+
+dict_file = os.listdir(in_file)
+
+def main(dict_file, in_file, out_file):
+    for file in tqdm(dict_file):
+        img = Image.open(in_file + file)
+        w,h = img.size
+        ratio = w/h
+        new_w = 640
+        new_h = int(new_w/ratio) 
+        new_img = img.resize((new_w,new_h), Image.ANTIALIAS)
+        file = file[:-4] + '.jpg'
+        new_img.save(out_file + file, quality = 100)
+
+if __name__ == '__main__':
+    main(dict_file, in_file, out_file)

result/best.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e817c0e164e80cd3cafb502ccb17d4ab2c98cfe18894980973a18d9b1b0ac58
+size 6210936

result/last.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7dc1bbb6403aa7a87940e46dab3f33aa4686b2d9013edc33243539d80ecc03bf
+size 6210936

result/result.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac418ee5c59d60e1c0b830ea592ac522b97c3158d0862b7cf25edcfb7c5bb49f
+size 40649916

result/result_2.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb70e20a7cd1f4de65b5b596de7416d435d31f8ba3e9f0ceaf0c4fe9315ca4f
+size 44911056

yolov8-pokemon-object-detection-main.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a1ce7dc3391a438c069135983d5ad12b3ea4066d9dbcbd0d14d25f3c24c58a2
+size 106562285