{ "cells": [ { "cell_type": "markdown", "metadata": { "colab_type": "text", "id": "EPcg39zEQY9T" }, "source": [ "# Yolo V3 Object Detection\n", "Notebook is copied from face_extraction.ipynb.\n", "\n", "Useful [resource](https://medium.com/@amrokamal_47691/yolo-yolov2-and-yolov3-all-you-want-to-know-7e3e92dc4899) on YOLO" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 34 }, "colab_type": "code", "executionInfo": { "elapsed": 3438, "status": "ok", "timestamp": 1590344047273, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "sNI9YJehDOYd", "outputId": "43712fc6-9f29-498a-e9ae-9b52657f719e" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Mounted at /content/drive\n" ] } ], "source": [ "from google.colab import drive\n", "drive.mount('/content/drive', force_remount=True)\n", "\n" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 187 }, "colab_type": "code", "executionInfo": { "elapsed": 4491, "status": "ok", "timestamp": 1590344048373, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "jP7qWA05EGQC", "outputId": "b6930023-a7ea-499a-a972-69c93837eae9" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "10_10.mp4\n", "101_10.mp4\n", "101_11.mp4\n", "101_12.mp4\n", "101_13.mp4\n", "101_14.mp4\n", "101_15.mp4\n", "101_16.mp4\n", "101_17.mp4\n", "101_18.mp4\n" ] } ], "source": [ "!ls drive/'My Drive'/cs231n-project/datasets/emotiw/train | head\n" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 136 }, "colab_type": "code", "executionInfo": { "elapsed": 10831, "status": "ok", "timestamp": 1590344054724, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "FA4epeBeIJY_", "outputId": "70069136-d227-4600-976f-630d10798cc6" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Requirement already satisfied: dlib in /usr/local/lib/python3.6/dist-packages (19.18.0)\n", "Requirement already satisfied: face_recognition in /usr/local/lib/python3.6/dist-packages (1.3.0)\n", "Requirement already satisfied: face-recognition-models>=0.3.0 in /usr/local/lib/python3.6/dist-packages (from face_recognition) (0.3.0)\n", "Requirement already satisfied: dlib>=19.7 in /usr/local/lib/python3.6/dist-packages (from face_recognition) (19.18.0)\n", "Requirement already satisfied: Pillow in /usr/local/lib/python3.6/dist-packages (from face_recognition) (7.0.0)\n", "Requirement already satisfied: numpy in /usr/local/lib/python3.6/dist-packages (from face_recognition) (1.18.4)\n", "Requirement already satisfied: Click>=6.0 in /usr/local/lib/python3.6/dist-packages (from face_recognition) (7.1.2)\n" ] } ], "source": [ "!pip3 install dlib\n", "!pip3 install face_recognition" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 34 }, "colab_type": "code", "executionInfo": { "elapsed": 14091, "status": "ok", "timestamp": 1590344057993, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "KICsoMbG2FuG", "outputId": "2b6bf400-d46e-465e-ce16-cc0fdc5d27a6" }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Using TensorFlow backend.\n" ] } ], "source": [ "import glob\n", "import json\n", "import math\n", "import os\n", "import pickle \n", "import sys\n", "\n", "import pprint\n", "\n", "# enter the foldername in your Drive where you have saved the project contents\n", "TRAINFOLDERNAME = 'datasets/emotiw/train'\n", "MODELFOLDERNAME = 'datasets/emotiw/models'\n", "assert TRAINFOLDERNAME is not None, \"[!] Enter the foldername.\"\n", "assert MODELFOLDERNAME is not None, \"[!] Enter the foldername.\"\n", "\n", "PROJECT_PATH = os.path.join('/content', 'drive', 'My Drive', 'cs231n-project')\n", "TRAIN_PATH = os.path.join(PROJECT_PATH, TRAINFOLDERNAME)\n", "MODEL_PATH = os.path.join(PROJECT_PATH, MODELFOLDERNAME)\n", "\n", "sys.path.append(TRAIN_PATH)\n", "\n", "from concurrent.futures import ThreadPoolExecutor\n", "import cv2\n", "import face_recognition\n", "from google.colab.patches import cv2_imshow\n", "from keras.models import load_model\n", "from keras.preprocessing.image import load_img\n", "from keras.preprocessing.image import img_to_array\n", "import numpy as np\n", "\n", "from yolov3 import yolo3_one_file_to_detect_them_all as yolo3_functions\n", "\n", "# Note: No trailing slash\n", "DOWNSAMPLE_RATIO = 1\n", "SAMPLE_EVERY = 10 # eg. if using 30 FPS and SAMPLE_EVERY=10, we would produce at 3 FPS\n" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "n0CRLSIMRTWe" }, "outputs": [], "source": [ "def create_yolov3_model():\n", " \"\"\"\n", " Create YOLO V3 Model using Pretrained weights\n", "\n", " Mostly following this tutorial: https://machinelearningmastery.com/how-to-perform-object-detection-with-yolov3-in-keras/\n", " \"\"\"\n", " print('Defining the YOLO v3 Model')\n", " model = yolo3_functions.make_yolov3_model()\n", "\n", " print('Loading the model weights')\n", " weight_file = os.path.join(TRAIN_PATH, 'yolov3', 'yolov3.weights')\n", " weight_reader = yolo3_functions.WeightReader(weight_file)\n", "\n", " # Next, call the load_weights() function of the WeightReader instance\n", " # passing in our defined Keras model to set the weights into the layers.\n", " print('Setting the model weights')\n", " weight_reader.load_weights(model)\n", "\n", " # Next, save this model to a Keras compatible .h5 model file ready for later use.\n", " print('Saving the model')\n", " model.save(os.path.join(MODEL_PATH, 'yolov3_model.h5'))\n" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "v_lGc7ItRzBs" }, "outputs": [], "source": [ "def process_video(video_dir, video_name, video_num, total_videos, model):\n", " \"\"\"\n", " Processes a video by extracting the faces from each frame and creating a \n", " list of list of faces which is then saved as a pickled object.\n", "\n", " NOTE: Faces are not guaranteed to be the same across frames.\n", " eg. 'face 1' in frame 1 may not be the same as 'face 1' in frame 2\n", "\n", " Pickle Object Format:\n", " [\n", " [\n", " [frame 1, face 1],\n", " [frame 1, face 2],\n", " [frame 1, face 3]\n", " ],\n", " [\n", " [frame 2, face 1],\n", " [frame 2, face 2]\n", " ],\n", " ...\n", " ]\n", "\n", " \"\"\"\n", " print(f\"Processing video {video_num}/{total_videos} with name {video_name}\\n\")\n", " vidcap = cv2.VideoCapture(os.path.join(video_dir, video_name))\n", "\n", " input_length = int(vidcap.get(cv2.CAP_PROP_FRAME_COUNT))\n", " frame_width = int(vidcap.get(cv2.CAP_PROP_FRAME_WIDTH))\n", " frame_height = int(vidcap.get(cv2.CAP_PROP_FRAME_HEIGHT))\n", " fps = int(vidcap.get(cv2.CAP_PROP_FPS))\n", "\n", " metadata = {}\n", " metadata[video_name] = {}\n", " faces_all_frames = []\n", " success, orig_image = vidcap.read()\n", " count = 0\n", " i = 1\n", " while success:\n", " if count % SAMPLE_EVERY == 0:\n", " print(f\" Working on {video_name} frame {count} video num {video_num}\")\n", " # NOTE:\n", " # cv2_imshow is used by Colab in place of cv2.imshow('image', image) as \n", " # the latter command is known to crash Jupyter sessions in Colab\n", "\n", " if video_num == 0 and count == 0:\n", " print(f\"Showing sample frame for {video_name} with shape {orig_image.shape}\")\n", " print(\"=\" * 20)\n", "\n", " image_h, image_w = orig_image.shape[:2]\n", " print('printing type image')\n", " print(type(orig_image))\n", "\n", " if video_num == 0 and count == 0:\n", " print(\"Original frame: \")\n", " cv2_imshow(orig_image)\n", " print(\"\")\n", "\n", " # Convert from BGR color (OpenCV) to RGB color (face_recognition)\n", " # rgb_image = image[:, :, ::-1]\n", "\n", " # https://machinelearningmastery.com/how-to-perform-object-detection-with-yolov3-in-keras/\n", " # The first step is to load the Keras model. This might be the slowest part of making a prediction.\n", " # Resize the image because YOLO expects 416 by 416\n", " # define the expected input shape for the model\n", " input_w, input_h = 416, 416\n", " image = cv2.resize(orig_image, (input_w, input_h), interpolation = cv2.INTER_CUBIC)\n", " print('Shape of image is {}'.format(image.shape))\n", "\n", " # scale pixel values to [0, 1]\n", " image = image.astype('float32')\n", " image /= 255.0\n", "\n", " # add a dimension so that we have one sample\n", " image = np.expand_dims(image, 0)\n", "\n", " # We can now feed the photo into the Keras model and make a prediction.\n", " # Make Prediction\n", " yhat = model.predict(image)\n", "\n", " print('Summarize the shape of the list of arrays')\n", " print([a.shape for a in yhat])\n", "\n", " # define the anchors\n", " anchors = [[116,90, 156,198, 373,326], [30,61, 62,45, 59,119], [10,13, 16,30, 33,23]]\n", " \n", " # define the probability threshold for detected objects\n", " class_threshold = 0.6\n", " boxes = list()\n", " for i in range(len(yhat)):\n", " # decode the output of the network\n", " boxes += yolo3_functions.decode_netout(yhat[i][0], anchors[i], class_threshold, input_h, input_w)\n", " \n", " # correct the sizes of the bounding boxes for the shape of the image\n", " yolo3_functions.correct_yolo_boxes(boxes, image_h=image_h, image_w=image_w, net_h=input_h, net_w=input_w)\n", " \n", " # suppress non-maximal boxes\n", " yolo3_functions.do_nms(boxes, 0.5)\n", " \n", " # define the labels\n", " # Double check all the categories that YoloV3 can predict \n", " labels = [\"person\", \"bicycle\", \"car\", \"motorbike\", \"aeroplane\", \"bus\", \"train\", \"truck\",\n", " \"boat\", \"traffic light\", \"fire hydrant\", \"stop sign\", \"parking meter\", \"bench\",\n", " \"bird\", \"cat\", \"dog\", \"horse\", \"sheep\", \"cow\", \"elephant\", \"bear\", \"zebra\", \"giraffe\",\n", " \"backpack\", \"umbrella\", \"handbag\", \"tie\", \"suitcase\", \"frisbee\", \"skis\", \"snowboard\",\n", " \"sports ball\", \"kite\", \"baseball bat\", \"baseball glove\", \"skateboard\", \"surfboard\",\n", " \"tennis racket\", \"bottle\", \"wine glass\", \"cup\", \"fork\", \"knife\", \"spoon\", \"bowl\", \"banana\",\n", " \"apple\", \"sandwich\", \"orange\", \"broccoli\", \"carrot\", \"hot dog\", \"pizza\", \"donut\", \"cake\",\n", " \"chair\", \"sofa\", \"pottedplant\", \"bed\", \"diningtable\", \"toilet\", \"tvmonitor\", \"laptop\", \"mouse\",\n", " \"remote\", \"keyboard\", \"cell phone\", \"microwave\", \"oven\", \"toaster\", \"sink\", \"refrigerator\",\n", " \"book\", \"clock\", \"vase\", \"scissors\", \"teddy bear\", \"hair drier\", \"toothbrush\"]\n", " \n", " # get the details of the detected objects\n", " v_boxes, v_labels, v_scores = yolo3_functions.get_boxes(boxes, labels, class_threshold)\n", " \n", " # summarize what we found\n", " for i in range(len(v_boxes)):\n", " print(v_labels[i], v_scores[i])\n", " \n", " # draw what we found\n", " yolo3_functions.draw_boxes(orig_image, v_boxes, v_labels, v_scores)\n", "\n", " v_box_locations = [(v_box.xmin, v_box.xmax, v_box.ymin, v_box.ymax) for v_box in v_boxes]\n", "\n", " metadata[video_name][count] = {\n", " 'vboxes': v_box_locations,\n", " 'v_labels': v_labels,\n", " 'v_scores': v_scores,\n", " }\n", "\n", " success, orig_image = vidcap.read()\n", " count += 1\n", " video_num += 1\n", " vidcap.release()\n", "\n", " return metadata\n" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "colab_type": "code", "executionInfo": { "elapsed": 24767, "status": "ok", "timestamp": 1590344068698, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "KVEtW75SR44c", "outputId": "6ab17b38-5543-4ec7-fa94-9ce46b449c76" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Defining the YOLO v3 Model\n", "Loading the model weights\n", "Setting the model weights\n", "loading weights of convolution #0\n", "loading weights of convolution #1\n", "loading weights of convolution #2\n", "loading weights of convolution #3\n", "no convolution #4\n", "loading weights of convolution #5\n", "loading weights of convolution #6\n", "loading weights of convolution #7\n", "no convolution #8\n", "loading weights of convolution #9\n", "loading weights of convolution #10\n", "no convolution #11\n", "loading weights of convolution #12\n", "loading weights of convolution #13\n", "loading weights of convolution #14\n", "no convolution #15\n", "loading weights of convolution #16\n", "loading weights of convolution #17\n", "no convolution #18\n", "loading weights of convolution #19\n", "loading weights of convolution #20\n", "no convolution #21\n", "loading weights of convolution #22\n", "loading weights of convolution #23\n", "no convolution #24\n", "loading weights of convolution #25\n", "loading weights of convolution #26\n", "no convolution #27\n", "loading weights of convolution #28\n", "loading weights of convolution #29\n", "no convolution #30\n", "loading weights of convolution #31\n", "loading weights of convolution #32\n", "no convolution #33\n", "loading weights of convolution #34\n", "loading weights of convolution #35\n", "no convolution #36\n", "loading weights of convolution #37\n", "loading weights of convolution #38\n", "loading weights of convolution #39\n", "no convolution #40\n", "loading weights of convolution #41\n", "loading weights of convolution #42\n", "no convolution #43\n", "loading weights of convolution #44\n", "loading weights of convolution #45\n", "no convolution #46\n", "loading weights of convolution #47\n", "loading weights of convolution #48\n", "no convolution #49\n", "loading weights of convolution #50\n", "loading weights of convolution #51\n", "no convolution #52\n", "loading weights of convolution #53\n", "loading weights of convolution #54\n", "no convolution #55\n", "loading weights of convolution #56\n", "loading weights of convolution #57\n", "no convolution #58\n", "loading weights of convolution #59\n", "loading weights of convolution #60\n", "no convolution #61\n", "loading weights of convolution #62\n", "loading weights of convolution #63\n", "loading weights of convolution #64\n", "no convolution #65\n", "loading weights of convolution #66\n", "loading weights of convolution #67\n", "no convolution #68\n", "loading weights of convolution #69\n", "loading weights of convolution #70\n", "no convolution #71\n", "loading weights of convolution #72\n", "loading weights of convolution #73\n", "no convolution #74\n", "loading weights of convolution #75\n", "loading weights of convolution #76\n", "loading weights of convolution #77\n", "loading weights of convolution #78\n", "loading weights of convolution #79\n", "loading weights of convolution #80\n", "loading weights of convolution #81\n", "no convolution #82\n", "no convolution #83\n", "loading weights of convolution #84\n", "no convolution #85\n", "no convolution #86\n", "loading weights of convolution #87\n", "loading weights of convolution #88\n", "loading weights of convolution #89\n", "loading weights of convolution #90\n", "loading weights of convolution #91\n", "loading weights of convolution #92\n", "loading weights of convolution #93\n", "no convolution #94\n", "no convolution #95\n", "loading weights of convolution #96\n", "no convolution #97\n", "no convolution #98\n", "loading weights of convolution #99\n", "loading weights of convolution #100\n", "loading weights of convolution #101\n", "loading weights of convolution #102\n", "loading weights of convolution #103\n", "loading weights of convolution #104\n", "loading weights of convolution #105\n", "Saving the model\n" ] } ], "source": [ "create_yolov3_model()" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000, "output_embedded_package_id": "1HmVwdpkJ7nAJsBJbFcm7BM2Ef4TTbWry" }, "colab_type": "code", "executionInfo": { "elapsed": 100582, "status": "ok", "timestamp": 1590344144524, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "pYzgVGo8SlxZ", "outputId": "5b6f408c-153e-4363-fe85-6501c6d5549f" }, "outputs": [ { "data": { "text/plain": [ "Output hidden; open in https://colab.research.google.com to view." ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Testing for one video\n", "# Note you might get an error that says UserWarning: No training configuration found in save file\n", "# Seems like it's ok to ignore for now: https://github.com/fizyr/keras-retinanet/issues/207\n", "print('Loading YOLO v3 Keras Model')\n", "model = load_model(os.path.join(MODEL_PATH, 'yolov3_model.h5'))\n", "metadata = process_video(video_dir=TRAIN_PATH, video_name=\"2_1.mp4\", video_num=0, total_videos=10, model=model)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "colab_type": "code", "executionInfo": { "elapsed": 100705, "status": "ok", "timestamp": 1590344144662, "user": { "displayName": "Vincent La", "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GgURuojY3AHW_TPmPkZW4KsBTfKl2ldHQY4Lb0=s64", "userId": "02730725867920307275" }, "user_tz": 420 }, "id": "SSdynIqMspTX", "outputId": "014d106d-d863-4292-ff6f-e7d3339337b0" }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'2_1.mp4': {0: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [92.55717992782593,\n", " 99.32825565338135,\n", " 96.99269533157349,\n", " 96.49882912635803,\n", " 99.48840141296387,\n", " 82.02412128448486,\n", " 72.56042957305908,\n", " 80.83866834640503],\n", " 'vboxes': [(0, 123, 46, 334),\n", " (191, 383, 18, 379),\n", " (333, 458, 30, 374),\n", " (388, 492, 20, 402),\n", " (102, 229, 84, 352),\n", " (56, 441, 11, 447),\n", " (188, 242, 6, 157),\n", " (85, 155, 66, 239)]},\n", " 10: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor'],\n", " 'v_scores': [97.35585451126099,\n", " 89.28497433662415,\n", " 99.17716979980469,\n", " 94.09701228141785,\n", " 93.08160543441772,\n", " 73.23048710823059],\n", " 'vboxes': [(323, 452, 35, 326),\n", " (0, 113, 53, 333),\n", " (186, 379, 15, 375),\n", " (385, 493, 22, 399),\n", " (104, 220, 76, 364),\n", " (54, 436, 13, 447)]},\n", " 20: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor'],\n", " 'v_scores': [82.79736638069153,\n", " 73.54065775871277,\n", " 93.23654770851135,\n", " 99.30883646011353,\n", " 94.15586590766907,\n", " 78.61786484718323],\n", " 'vboxes': [(312, 428, 28, 326),\n", " (1, 106, 47, 333),\n", " (85, 199, 58, 344),\n", " (184, 370, 8, 383),\n", " (372, 493, 18, 400),\n", " (51, 439, 9, 456)]},\n", " 30: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor'],\n", " 'v_scores': [85.04635691642761,\n", " 88.33048939704895,\n", " 96.26598358154297,\n", " 99.44911003112793,\n", " 91.30313992500305,\n", " 66.667640209198],\n", " 'vboxes': [(298, 430, 21, 333),\n", " (-1, 116, 38, 348),\n", " (84, 202, 53, 346),\n", " (174, 368, 10, 382),\n", " (367, 493, 24, 392),\n", " (63, 424, 0, 461)]},\n", " 40: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [91.82323217391968,\n", " 95.71577310562134,\n", " 98.67913126945496,\n", " 99.55897927284241,\n", " 95.20530104637146,\n", " 96.48763537406921],\n", " 'vboxes': [(289, 438, 15, 342),\n", " (0, 96, 43, 347),\n", " (84, 208, 56, 348),\n", " (150, 359, 25, 364),\n", " (367, 494, 22, 396),\n", " (66, 139, 50, 218)]},\n", " 50: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [90.0563657283783,\n", " 99.60424304008484,\n", " 99.35174584388733,\n", " 91.84811115264893,\n", " 96.36315703392029,\n", " 71.69594168663025,\n", " 67.21392869949341,\n", " 61.619603633880615],\n", " 'vboxes': [(1, 108, 46, 335),\n", " (84, 204, 58, 347),\n", " (154, 354, 32, 364),\n", " (297, 430, 17, 375),\n", " (369, 493, 15, 399),\n", " (77, 408, 12, 450),\n", " (284, 347, 25, 153),\n", " (66, 138, 39, 311)]},\n", " 60: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [96.70511484146118,\n", " 65.13259410858154,\n", " 99.45477843284607,\n", " 99.14534091949463,\n", " 91.66932106018066,\n", " 96.00441455841064,\n", " 64.59804177284241,\n", " 71.03583812713623,\n", " 74.38929080963135],\n", " 'vboxes': [(0, 101, 42, 335),\n", " (41, 144, 62, 312),\n", " (80, 208, 60, 344),\n", " (152, 352, 33, 362),\n", " (294, 431, 16, 377),\n", " (369, 494, 13, 402),\n", " (74, 413, 10, 452),\n", " (165, 217, 0, 87),\n", " (279, 352, 28, 152)]},\n", " 70: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person'],\n", " 'v_scores': [81.25280141830444,\n", " 97.69538640975952,\n", " 97.94414043426514,\n", " 99.3894875049591,\n", " 97.39837050437927,\n", " 95.32923102378845,\n", " 60.59151887893677,\n", " 81.2296986579895],\n", " 'vboxes': [(1, 109, 43, 337),\n", " (28, 154, 60, 330),\n", " (102, 218, 60, 346),\n", " (150, 355, 32, 364),\n", " (306, 430, 31, 363),\n", " (371, 494, 19, 398),\n", " (71, 410, 4, 459),\n", " (285, 347, 31, 173)]},\n", " 80: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person'],\n", " 'v_scores': [77.4164080619812,\n", " 91.72790050506592,\n", " 99.61099624633789,\n", " 96.73118591308594,\n", " 95.14124393463135,\n", " 98.1213390827179,\n", " 62.19985485076904,\n", " 87.11501359939575],\n", " 'vboxes': [(4, 102, 28, 344),\n", " (45, 153, 61, 332),\n", " (140, 364, 30, 370),\n", " (307, 429, 27, 367),\n", " (372, 494, 9, 404),\n", " (105, 219, 66, 365),\n", " (78, 404, 10, 450),\n", " (287, 346, 34, 168)]},\n", " 90: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person'],\n", " 'v_scores': [96.33894562721252,\n", " 88.39572072029114,\n", " 77.95645594596863,\n", " 99.4691789150238,\n", " 93.85527968406677,\n", " 97.96686172485352,\n", " 83.637136220932,\n", " 77.89674997329712],\n", " 'vboxes': [(307, 423, 38, 310),\n", " (0, 103, 44, 340),\n", " (40, 153, 67, 322),\n", " (175, 374, 20, 376),\n", " (367, 494, 10, 403),\n", " (75, 210, 63, 361),\n", " (64, 424, 13, 449),\n", " (294, 343, 31, 151)]},\n", " 100: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'tvmonitor',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [97.08964824676514,\n", " 98.54891896247864,\n", " 99.57587718963623,\n", " 99.57835078239441,\n", " 92.68172979354858,\n", " 68.18132400512695,\n", " 61.17677688598633,\n", " 67.68649816513062],\n", " 'vboxes': [(302, 435, 28, 331),\n", " (0, 107, 52, 335),\n", " (84, 203, 55, 347),\n", " (163, 386, 27, 371),\n", " (384, 493, 14, 400),\n", " (113, 443, 19, 446),\n", " (172, 231, 3, 163),\n", " (73, 143, 57, 214)]},\n", " 110: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [97.6650059223175,\n", " 99.16821718215942,\n", " 99.07457828521729,\n", " 99.13824200630188,\n", " 63.58157992362976,\n", " 75.685715675354,\n", " 77.40837931632996],\n", " 'vboxes': [(1, 110, 54, 339),\n", " (87, 198, 57, 345),\n", " (141, 370, 24, 365),\n", " (287, 458, 27, 370),\n", " (169, 224, 9, 159),\n", " (84, 152, 58, 215),\n", " (431, 491, 40, 370)]},\n", " 120: {'v_labels': ['person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person',\n", " 'person'],\n", " 'v_scores': [99.42789673805237,\n", " 89.81438279151917,\n", " 98.99359941482544,\n", " 86.90922856330872,\n", " 80.98921179771423,\n", " 70.37965059280396],\n", " 'vboxes': [(284, 488, 10, 361),\n", " (0, 128, 24, 347),\n", " (133, 285, 50, 348),\n", " (418, 480, 34, 114),\n", " (265, 319, 62, 193),\n", " (98, 182, 46, 230)]}}}\n" ] } ], "source": [ "pprint.pprint(metadata)" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 1000 }, "colab_type": "code", "id": "8OdoPISG6HaB", "outputId": "bb103809-d80d-4de1-ea89-37b9a9dd1c71" }, "outputs": [], "source": [ "# Running over all videos now\n", "\n", "# Getting a list of all video files in a directory\n", "file_path_names = glob.glob(os.path.join(TRAIN_PATH, '*.mp4'))\n", "filenames = [os.path.basename(file_path_name) for file_path_name in file_path_names]\n", "\n", "# # Just testing a few, can comment out later\n", "# filenames = ['344_7.mp4', '260_13.mp4', '220_4.mp4', '245_8.mp4']\n", "\n", "print(filenames)\n", "\n", "# Process all videos, run YOLO, and store output\n", "metadata_all = {}\n", "for video_filename in filenames:\n", " print('Processing video: {}'.format(video_filename))\n", " metadata = process_video(video_dir=TRAIN_PATH, video_name=video_filename, video_num=0, total_videos=10, model=model)\n", " metadata_all.update(metadata)" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "CZNl1GLMCUxZ" }, "outputs": [], "source": [ "with open(os.path.join(TRAIN_PATH, 'yolov3', 'metadata_all.json'), 'w') as fp:\n", " json.dump(metadata_all, fp, sort_keys=True, indent=4, separators=(',', ': '))" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "Jwvn0TyM3NEs" }, "outputs": [], "source": [] }, { "cell_type": "code", "execution_count": 0, "metadata": { "colab": {}, "colab_type": "code", "id": "31piHUFySjX3" }, "outputs": [], "source": [] } ], "metadata": { "accelerator": "GPU", "colab": { "collapsed_sections": [], "name": "yolov3_object_detection.ipynb", "provenance": [] }, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.9" } }, "nbformat": 4, "nbformat_minor": 1 }