Merge branch 'main' of https://github.com/vjshankar4204/MediaPipePose…

…Estimation-eschar

SquatMaster/ML_training/test1/ML_training_test1.py

@@ -1,24 +1,102 @@
-import matplotlib.pyplot as plt
+import cv2
+import mediapipe as mp
 import numpy as np
-import PIL
-import tensorflow as tf
+import sys
+sys.path.insert(0,'../functions')
+mp_drawing = mp.solutions.drawing_utils
+mp_pose = mp.solutions.pose
+import os
 
-from tensorflow import keras
-from tensorflow.keras import layers
-from tensorflow.keras.models import Sequential
+# Specify the location of the file with the video to be read
+# Capture frames from a saved video
+filename_r = '../../Data/videos/exercise_stock_video3.mp4'
+# filename_r = 4
+# Specify the location where the new video witwwh detections will be written
+# filename_w = '../Data/videos/exercise_stock_video3_wRepCount.mp4'
 
-import pathlib
+cap = cv2.VideoCapture(filename_r)
 
-dataset_url = "https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz"
-data_dir = tf.keras.utils.get_file('flower_photos.tar', origin=dataset_url, extract=True)
-data_dir = pathlib.Path(data_dir).with_suffix('')
+# Get video frame dimensions and fps
+frame_width = int(cap.get(3))
+frame_height = int(cap.get(4))
+frame_size = (frame_width,frame_height)
+fps = cap.get(5)
 
-image_count = len(list(data_dir.glob('*/*.jpg')))
-print(image_count)
+# # Create a video writer object
+# output = cv2.VideoWriter(filename_w, cv2.VideoWriter_fourcc(*'XVID'), fps, frame_size)
 
+frame_count = 0
+
+## Setup mediapipe instance
+with mp_pose.Pose(min_detection_confidence=0.9, min_tracking_confidence=0.5, enable_segmentation=True) as pose:
+
+    fps = cap.get(5)
+    frame_width  = int(cap.get(3))
+    frame_height = int(cap.get(4))
+
+    while cap.isOpened():
+        ret, frame = cap.read()
+
+        # Stop the code when the video is over
+        if not ret:
+            print('Failed to grab a frame')
+            break
+
+        # Count frames and calculate time in video
+        frame_count += 1
+        time = frame_count / fps #[s]
+
+        # Recolor image to RGB
+        image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+        image.flags.writeable = False
+
+        # Make detection
+        results = pose.process(image)
+
+        # Recolor back to BGR
+        image.flags.writeable = True
+        image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+
+        # Extract landmarks
+        try:
+            landmarks = results.pose_landmarks.landmark
+
+            # Get coordinates
+            hip = [landmarks[mp_pose.PoseLandmark.RIGHT_HIP.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_HIP.value].y]
+            knee = [landmarks[mp_pose.PoseLandmark.RIGHT_KNEE.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_KNEE.value].y]
+            ankle = [landmarks[mp_pose.PoseLandmark.RIGHT_ANKLE.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_ANKLE.value].y]
+
+
+
+        except:
+            pass
+
+
+
+        # Render detections
+        mp_drawing.draw_landmarks(image, results.pose_landmarks, mp_pose.POSE_CONNECTIONS,
+                                mp_drawing.DrawingSpec(color=(245,117,66), thickness=2, circle_radius=2), 
+                                mp_drawing.DrawingSpec(color=(245,66,230), thickness=2, circle_radius=2) 
+                                 )
+
+
+        # Display and write the video
+        if ret == True:
+            cv2.imshow('Mediapipe Feed', image)
+            # output.write(image)
+
+            # Pause or stop the video when instructed
+            key = cv2.waitKey(5)
+            # Stop by pressing 'q'
+            if key == ord('q'):
+                break
+            # Pause by pressing 'w', resume by pressing any other key
+            if key == ord('w'):
+                cv2.waitKey(-1)
+
+
+    cap.release()
+    cv2.destroyAllWindows()
 
 
-roses = list(data_dir.glob('roses/*'))
-im = PIL.Image.open(str(roses[1]))
 
-im.show()

SquatMaster/ML_training/test1/ML_training_test2.ipynb

@@ -0,0 +1,194 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 61,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import cv2\n",
+    "import mediapipe as mp\n",
+    "import numpy as np\n",
+    "import sys\n",
+    "sys.path.insert(0,'../../functions')\n",
+    "mp_drawing = mp.solutions.drawing_utils\n",
+    "mp_pose = mp.solutions.pose\n",
+    "import os\n",
+    "from heel_crop1 import heel_crop\n",
+    "\n",
+    "# from tensorflow import keras\n",
+    "# from tensorflow.keras import layers\n",
+    "# from tensorflow.keras.models import Sequential"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "I0000 00:00:1701556954.369788  158802 gl_context_egl.cc:85] Successfully initialized EGL. Major : 1 Minor: 5\n",
+      "I0000 00:00:1701556954.371567  192045 gl_context.cc:344] GL version: 3.2 (OpenGL ES 3.2 Mesa 23.2.1-1ubuntu3), renderer: AMD Radeon Graphics (renoir, LLVM 15.0.7, DRM 3.54, 6.5.0-13-generic)\n",
+      "qt.qpa.xcb: QXcbConnection: XCB error: 146 (Unknown), sequence: 38249, resource id: 0, major code: 139 (Unknown), minor code: 20\n",
+      "qt.qpa.xcb: QXcbConnection: XCB error: 146 (Unknown), sequence: 38254, resource id: 1329, major code: 139 (Unknown), minor code: 20\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "# Specify the location of the file with the video to be read\n",
+    "# Capture frames from a saved video\n",
+    "filename_r = '../../Data/videos/exercise_stock_video3.mp4'\n",
+    "# filename_r = 4\n",
+    "filename_r = '../../../../Material/hill_raise/IMG_3719.MOV'\n",
+    "# Specify the location where the new video with detections will be written\n",
+    "# filename_w = '../Data/videos/exercise_stock_video3_wRepCount.mp4'\n",
+    "\n",
+    "cap = cv2.VideoCapture(filename_r)\n",
+    "\n",
+    "# Get video frame dimensions and fps\n",
+    "frame_width = int(cap.get(3))\n",
+    "frame_height = int(cap.get(4))\n",
+    "frame_size = (frame_width,frame_height)\n",
+    "fps = cap.get(5)\n",
+    "\n",
+    "# # Create a video writer object\n",
+    "# output = cv2.VideoWriter(filename_w, cv2.VideoWriter_fourcc(*'XVID'), fps, frame_size)\n",
+    "\n",
+    "frame_count = 0\n",
+    "frame_count_trim = 0\n",
+    "\n",
+    "## Setup mediapipe instance\n",
+    "with mp_pose.Pose(min_detection_confidence=0.9, min_tracking_confidence=0.5, enable_segmentation=True) as pose:\n",
+    "\n",
+    "    fps = cap.get(5)\n",
+    "    frame_width  = int(cap.get(3))\n",
+    "    frame_height = int(cap.get(4))\n",
+    "\n",
+    "    while cap.isOpened():\n",
+    "        ret, frame = cap.read()\n",
+    "\n",
+    "        # Rotate the video feed if its coming from the phone webcam\n",
+    "        if filename_r == 4:\n",
+    "            frame=cv2.rotate(frame, cv2.ROTATE_90_CLOCKWISE)\n",
+    "\n",
+    "\n",
+    "        # Stop the code when the video is over\n",
+    "        if not ret:\n",
+    "            print('Failed to grab a frame')\n",
+    "            break\n",
+    "\n",
+    "        # Count frames and calculate time in video\n",
+    "        frame_count += 1\n",
+    "        time = frame_count / fps #[s]\n",
+    "\n",
+    "        # Recolor image to RGB\n",
+    "        image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)\n",
+    "        image.flags.writeable = False\n",
+    "      \n",
+    "        # Make detection\n",
+    "        results = pose.process(image)\n",
+    "    \n",
+    "        # Recolor back to BGR\n",
+    "        image.flags.writeable = True\n",
+    "        image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)\n",
+    "\n",
+    "        image_trimmed = frame\n",
+    "\n",
+    "        # Extract landmarks\n",
+    "        try:\n",
+    "            landmarks = results.pose_landmarks.landmark\n",
+    "            \n",
+    "            # Get coordinates\n",
+    "            right_ear = [landmarks[mp_pose.PoseLandmark.RIGHT_EAR.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_EAR.value].y]\n",
+    "            left_ankle = [landmarks[mp_pose.PoseLandmark.LEFT_ANKLE.value].x,landmarks[mp_pose.PoseLandmark.LEFT_ANKLE.value].y]\n",
+    "            right_ankle = [landmarks[mp_pose.PoseLandmark.RIGHT_ANKLE.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_ANKLE.value].y]\n",
+    "            left_heel = [landmarks[mp_pose.PoseLandmark.LEFT_HEEL.value].x,landmarks[mp_pose.PoseLandmark.LEFT_HEEL.value].y]\n",
+    "            right_heel = [landmarks[mp_pose.PoseLandmark.RIGHT_HEEL.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_HEEL.value].y]\n",
+    "            left_footIndex = [landmarks[mp_pose.PoseLandmark.LEFT_FOOT_INDEX.value].x,landmarks[mp_pose.PoseLandmark.LEFT_FOOT_INDEX.value].y]\n",
+    "            right_footIndex = [landmarks[mp_pose.PoseLandmark.RIGHT_FOOT_INDEX.value].x,landmarks[mp_pose.PoseLandmark.RIGHT_FOOT_INDEX.value].y]\n",
+    "            \n",
+    "        \n",
+    "            points = [left_ankle,left_heel,left_footIndex]\n",
+    "            [image_trimmed, frame_count_trim] = heel_crop(points, frame, frame_count_trim)\n",
+    "            \n",
+    "        except:\n",
+    "            pass\n",
+    "        \n",
+    "\n",
+    "        # Render detections\n",
+    "        mp_drawing.draw_landmarks(image, results.pose_landmarks, mp_pose.POSE_CONNECTIONS,\n",
+    "                                mp_drawing.DrawingSpec(color=(245,117,66), thickness=2, circle_radius=2), \n",
+    "                                mp_drawing.DrawingSpec(color=(245,66,230), thickness=2, circle_radius=2) \n",
+    "                                 )\n",
+    "\n",
+    "\n",
+    "        # Display and write the video\n",
+    "        if ret == True:\n",
+    "            cv2.imshow('Mediapipe Feed', image)\n",
+    "            # output.write(image)\n",
+    "\n",
+    "            try:\n",
+    "                cv2.imshow('Right foot', image_trimmed)\n",
+    "                # if frame_count_trim % 5 == 0:\n",
+    "                    # cv2.imwrite(f'/home/rafik/PROJECTS/pose1/Material/hill_raise/cropped_feet/left_foot/left_foot_{frame_count_trim}.jpg', image_trimmed)\n",
+    "            except:\n",
+    "                print('No foot in sight')\n",
+    "\n",
+    "            # Pause or stop the video when instructed\n",
+    "            key = cv2.waitKey(5)\n",
+    "            # Stop by pressing 'q'\n",
+    "            if key == ord('q'):\n",
+    "                break\n",
+    "            # Pause by pressing 'w', resume by pressing any other key\n",
+    "            if key == ord('w'):\n",
+    "                cv2.waitKey(-1)\n",
+    "\n",
+    "\n",
+    "    cap.release()\n",
+    "    cv2.destroyAllWindows()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 56,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "/home/rafik/PROJECTS/pose1/MediaPipePoseEstimation-eschar/SquatMaster/ML_training/test1\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(os.getcwd())"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "venv_pose1",
+   "language": "python",
+   "name": "venv_pose1"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}