initial commit

main.py

@@ -0,0 +1,20 @@
+from ultralytics import YOLO
+
+# Load a pretrained YOLO11n model
+model = YOLO("yolo11m-2_uniform.onnx")
+model_pose = YOLO("yolo11s-pose.pt")
+
+# Define path to video file
+source = "PF-071124-2.mp4"
+
+# Run inference on the source
+results = model(source, stream=True)  # generator of Results objects
+
+for result in results:
+    boxes = result.boxes  # Boxes object for bounding box outputs
+    masks = result.masks  # Masks object for segmentation masks outputs
+    keypoints = result.keypoints  # Keypoints object for pose outputs
+    probs = result.probs  # Probs object for classification outputs
+    obb = result.obb  # Oriented boxes object for OBB outputs
+    result.show()  # display to screen
+    result.save(filename="result.jpg")  # save to disk

main2.py

@@ -0,0 +1,37 @@
+from ultralytics import YOLO
+import cv2
+import os
+import numpy as np
+
+# Load a pretrained YOLO11n model
+model = YOLO("yolo11m-2_uniform.onnx")
+model_pose = YOLO("yolo11s-pose.pt")
+
+video_path = "PF-071124-2.mp4"
+cap = cv2.VideoCapture(video_path)
+
+ret = True
+frame_count = 0  # Initialize a frame counter
+save_dir = "/results/"  # Directory to save the crops
+os.makedirs(save_dir, exist_ok=True)  # Create the directory if it doesn't exist
+
+while ret:
+    ret, frame = cap.read()
+    frame_count += 1  # Increment the frame counter
+    #frame2_, frame_ = np.empty(2)
+    if ret:
+        results = model.predict(frame, conf=0.5)
+        frame_ = results[0].plot()
+        pose_results = model_pose.predict(frame_, show_boxes=False, show_conf=False, show_labels=False)
+        results[0].keypoints = pose_results[0].keypoints
+
+        frame_ = results[0].plot()
+
+        
+        cv2.imshow("frame", frame_)
+            
+        if cv2.waitKey(25) & 0xFF == ord('q'):
+            break
+
+cap.release()
+cv2.destroyAllWindows()

main3.py

@@ -0,0 +1,50 @@
+from ultralytics import YOLO
+import cv2
+import os
+import numpy as np
+
+# Load the object detection model
+det_model = YOLO("yolo11m-2_uniform.onnx")  # General object detection model
+
+# Load the pose estimation model
+pose_model = YOLO("yolo11s-pose.pt")  # Pose estimation model
+
+video_path = "PF-071124-2.mp4"
+cap = cv2.VideoCapture(video_path)
+
+ret = True
+frame_count = 0  # Initialize a frame counter
+save_dir = "/results/"  # Directory to save the crops
+os.makedirs(save_dir, exist_ok=True)  # Create the directory if it doesn't exist
+
+while ret:
+    ret, frame = cap.read()
+    frame_count += 1  # Increment the frame counter
+    frame2_, frame_ = np.empty(2)
+    if ret:
+        # Run object detection
+        det_results = det_model.predict(frame, conf=0.5)
+
+        # Filter detections for persons (class 0)
+        person_detections = [det for det in det_results if det.names[0] == 'crew']
+
+        #for i, det in enumerate(person_detections[0].boxes.xyxy):
+            # Extract bounding box coordinates
+        #    x1, y1, x2, y2 = map(int, det[:4])
+        #    crop = frame[y1:y2, x1:x2] 
+        
+        # Run pose estimation on detected persons
+        for i, person in person_detections:
+            #how to return the tensor to posemodel???
+            x1, y1, x2, y2 = map(int, person[i].boxes.xyxy[:4])
+            person_image = frame[y1:y2, x1:x2]  # Crop the person from the image
+            
+            pose_results = pose_model(person_image)
+            frame_ = pose_results[0].plot()
+        cv2.imshow("frame", frame_)
+            
+        if cv2.waitKey(25) & 0xFF == ord('q'):
+            break
+
+cap.release()
+cv2.destroyAllWindows()