multi object tracking

.gitattributes

@@ -35,3 +35,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 files/b.png filter=lfs diff=lfs merge=lfs -text
 *.png filter=lfs diff=lfs merge=lfs -text
+*.MOV filter=lfs diff=lfs merge=lfs -text
+*.mp4 filter=lfs diff=lfs merge=lfs -text

app.py

@@ -6,6 +6,7 @@ Sometimes you just want to take an image from your database and see how it chang
 import gradio as gr
 from  src.threshold_methods import threshold_methods
 from src.yolo import yolo
+from src.video_processing import video_processing
 import cv2
 
 new_outputs = [
@@ -36,7 +37,7 @@ HuggingFace = gr.Interface(
 )
 
 gr.TabbedInterface(
-    [HuggingFace,threshold_methods,yolo],
-    tab_names=['HuggingFace','Thresholding Image Segmentation','YOLO']
+    [HuggingFace,threshold_methods,yolo,video_processing],
+    tab_names=['HuggingFace','Thresholding Image Segmentation','YOLO','video_processing']
 ).queue().launch()
 

files/a.MOV

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dddb5efbd2e9dd31904e9004a15de3a47e64efd6e5a164d7e7f3a66ec581633
+size 26720050

files/a.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92385551e0b46905515c6d06e61c1c727eebbd5ca48f8c05ac1f2f23c4cf83ed
+size 20323167

files/b.mov

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afe123619d4402adf016dc46838f66998d3971b8bf693af75fa48619ced2d65a
+size 4966448

requirements.txt

@@ -4,7 +4,8 @@ requests
 hydra-core
 ultralytics
 gradio
-
+ffmpeg
+ffmpeg-python
 
 # # Ultralytics requirements
 # # Usage: pip install -r requirements.txt

src/test.py

@@ -1,6 +1,12 @@
 from ultralytics import YOLO
 from PIL import Image
 import cv2
+import ffmpeg
+ffmpeg.input('files/a.MOV').output('files/a.mp4').run()
+ffmpeg.input('input.mov').output('output.mp4').run()
+
+
+
 def draw_boxes(image, boxes):
     for box in boxes:
         
@@ -24,9 +30,41 @@ def detect_objects_on_image(buf):
         ])
     return output
 
+# model = MaskRCNN("mask_rcnn_model.pth") 
+# results = model.predict(img)
+# masks = results['masks']
+
+# img = cv2.imread('a.png')   
+# boxes=detect_objects_on_image(img)
+# img_with_boxes = draw_boxes(img, boxes)
+# cv2.imshow("test",img_with_boxes)
+# cv2.waitKey(0)
+model = YOLO("files/yolov8n.pt")    
+video_path = "files/a.MOV"
+cap = cv2.VideoCapture(video_path)
+
+# Loop through the video frames
+while cap.isOpened():
+    # Read a frame from the video
+    success, frame = cap.read()
+
+    if success:
+        # Run YOLOv8 tracking on the frame, persisting tracks between frames
+        results = model.track(frame, persist=True)
+
+        # Visualize the results on the frame
+        annotated_frame = results[0].plot()
+
+        # Display the annotated frame
+        cv2.imshow("YOLOv8 Tracking", annotated_frame)
+
+        # Break the loop if 'q' is pressed
+        if cv2.waitKey(1) & 0xFF == ord("q"):
+            break
+    else:
+        # Break the loop if the end of the video is reached
+        break
 
-img = cv2.imread('a.png')   
-boxes=detect_objects_on_image(img)
-img_with_boxes = draw_boxes(img, boxes)
-cv2.imshow("test",img_with_boxes)
-cv2.waitKey(0)
+# Release the video capture object and close the display window
+cap.release()
+cv2.destroyAllWindows()

src/test2.py

@@ -0,0 +1 @@
+# www.pirahansiah.com

src/video_processing.py

@@ -0,0 +1,25 @@
+import cv2
+import numpy as np
+import gradio as gr
+from ultralytics import YOLO
+video_path = [['files/a.mp4']]
+inputs = [gr.components.Video(type="filepath", label="Input ")]
+outputs = [gr.components.Image(type="numpy", label="Output ")]
+model = YOLO("files/yolov8n.pt")   
+def show_preds_video(video_file):
+    cap = cv2.VideoCapture(video_file)         
+    while cap.isOpened():
+        success, frame = cap.read()      
+        if success:             
+            results = model.track(frame, persist=True)             
+            annotated_frame = results[0].plot()             
+            yield cv2.cvtColor(annotated_frame, cv2.COLOR_BGR2RGB)                        
+     
+video_processing = gr.Interface(
+    fn=show_preds_video,
+    inputs=inputs,
+    outputs=outputs,     
+    examples=video_path,
+    cache_examples=False,
+    title="Video Multi Objects tracking",    
+    )