In [1]:
!pip install tensorflow matplotlib pandas
Requirement already satisfied: tensorflow in /home/bhavika/.local/lib/python3.8/site-packages (2.3.1) Requirement already satisfied: matplotlib in /home/bhavika/.local/lib/python3.8/site-packages (3.3.2) Requirement already satisfied: pandas in /home/bhavika/.local/lib/python3.8/site-packages (1.3.0) Requirement already satisfied: gast==0.3.3 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (0.3.3) Requirement already satisfied: numpy<1.19.0,>=1.16.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (1.18.5) Requirement already satisfied: wheel>=0.26 in /usr/lib/python3/dist-packages (from tensorflow) (0.34.2) Requirement already satisfied: six>=1.12.0 in /usr/lib/python3/dist-packages (from tensorflow) (1.14.0) Requirement already satisfied: opt-einsum>=2.3.2 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (3.3.0) Requirement already satisfied: tensorflow-estimator<2.4.0,>=2.3.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (2.3.0) Requirement already satisfied: wrapt>=1.11.1 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (1.12.1) Requirement already satisfied: grpcio>=1.8.6 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (1.38.1) Requirement already satisfied: astunparse==1.6.3 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (1.6.3) Requirement already satisfied: h5py<2.11.0,>=2.10.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (2.10.0) Requirement already satisfied: protobuf>=3.9.2 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (3.13.0) Requirement already satisfied: keras-preprocessing<1.2,>=1.1.1 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (1.1.2) Requirement already satisfied: tensorboard<3,>=2.3.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (2.6.0) Requirement already satisfied: absl-py>=0.7.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (0.11.0) Requirement already satisfied: google-pasta>=0.1.8 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (0.2.0) Requirement already satisfied: termcolor>=1.1.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorflow) (1.1.0) Requirement already satisfied: pillow>=6.2.0 in /home/bhavika/.local/lib/python3.8/site-packages (from matplotlib) (6.2.2) Requirement already satisfied: kiwisolver>=1.0.1 in /home/bhavika/.local/lib/python3.8/site-packages (from matplotlib) (1.3.1) Requirement already satisfied: python-dateutil>=2.1 in /usr/lib/python3/dist-packages (from matplotlib) (2.7.3) Requirement already satisfied: certifi>=2020.06.20 in /home/bhavika/.local/lib/python3.8/site-packages (from matplotlib) (2020.6.20) Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3 in /home/bhavika/.local/lib/python3.8/site-packages (from matplotlib) (2.4.7) Requirement already satisfied: cycler>=0.10 in /home/bhavika/.local/lib/python3.8/site-packages (from matplotlib) (0.10.0) Requirement already satisfied: pytz>=2017.3 in /usr/lib/python3/dist-packages (from pandas) (2019.3) Requirement already satisfied: setuptools in /usr/lib/python3/dist-packages (from protobuf>=3.9.2->tensorflow) (45.2.0) Requirement already satisfied: tensorboard-data-server<0.7.0,>=0.6.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorboard<3,>=2.3.0->tensorflow) (0.6.1) Requirement already satisfied: requests<3,>=2.21.0 in /usr/lib/python3/dist-packages (from tensorboard<3,>=2.3.0->tensorflow) (2.22.0) Requirement already satisfied: google-auth<2,>=1.6.3 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorboard<3,>=2.3.0->tensorflow) (1.33.1) Requirement already satisfied: tensorboard-plugin-wit>=1.6.0 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorboard<3,>=2.3.0->tensorflow) (1.7.0) Requirement already satisfied: google-auth-oauthlib<0.5,>=0.4.1 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorboard<3,>=2.3.0->tensorflow) (0.4.2) Requirement already satisfied: werkzeug>=0.11.15 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorboard<3,>=2.3.0->tensorflow) (1.0.1) Requirement already satisfied: markdown>=2.6.8 in /home/bhavika/.local/lib/python3.8/site-packages (from tensorboard<3,>=2.3.0->tensorflow) (3.3.3) Requirement already satisfied: pyasn1-modules>=0.2.1 in /home/bhavika/.local/lib/python3.8/site-packages (from google-auth<2,>=1.6.3->tensorboard<3,>=2.3.0->tensorflow) (0.2.8) Requirement already satisfied: rsa<5,>=3.1.4; python_version >= "3.6" in /home/bhavika/.local/lib/python3.8/site-packages (from google-auth<2,>=1.6.3->tensorboard<3,>=2.3.0->tensorflow) (4.6) Requirement already satisfied: cachetools<5.0,>=2.0.0 in /home/bhavika/.local/lib/python3.8/site-packages (from google-auth<2,>=1.6.3->tensorboard<3,>=2.3.0->tensorflow) (4.1.1) Requirement already satisfied: requests-oauthlib>=0.7.0 in /home/bhavika/.local/lib/python3.8/site-packages (from google-auth-oauthlib<0.5,>=0.4.1->tensorboard<3,>=2.3.0->tensorflow) (1.3.0) Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in /home/bhavika/.local/lib/python3.8/site-packages (from pyasn1-modules>=0.2.1->google-auth<2,>=1.6.3->tensorboard<3,>=2.3.0->tensorflow) (0.4.8) Requirement already satisfied: oauthlib>=3.0.0 in /usr/lib/python3/dist-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<0.5,>=0.4.1->tensorboard<3,>=2.3.0->tensorflow) (3.1.0)
In [2]:
import tensorflow as tf
import numpy as np
from PIL import Image, ImageDraw, ImageOps
import matplotlib.pyplot as plt
import pandas as pd
import os
import cv2
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)
In [3]:
from src.pipeline.pose_engine import PoseEngine
from src.pipeline.fall_detect import FallDetector
In [4]:
def _fall_detect_config_movenet():
config = {
'model': {
'tflite': 'ai_models/lite-model_movenet_singlepose_thunder_3.tflite',
},
'labels': 'ai_models/pose_labels.txt',
'confidence_threshold': 0.6,
'model_name':'movenet'
}
return config
In [5]:
def _fall_detect_config_mobilenet():
config = {
'model': {
'tflite': 'ai_models/posenet_mobilenet_v1_100_257x257_multi_kpt_stripped.tflite',
},
'labels': 'ai_models/pose_labels.txt',
'confidence_threshold': 0.6,
'model_name':'mobilenet'
}
return config
In [6]:
movenet_config = _fall_detect_config_movenet()
movenet_fall_detector = FallDetector(**movenet_config)
movenet_pose_engine = PoseEngine(movenet_fall_detector._tfengine, movenet_config['model_name'])
mobilenet_config = _fall_detect_config_mobilenet()
mobilenet_fall_detector = FallDetector(**mobilenet_config)
mobilenet_pose_engine = PoseEngine(mobilenet_fall_detector._tfengine, mobilenet_config['model_name'])
Posenet Vs Movenet¶
In [7]:
import os
DATADIR = 'Images'
comparison_df = pd.DataFrame()
for template_path in os.listdir(DATADIR):
template_image_src = Image.open(os.path.join(DATADIR,template_path))
_thumbnail, movenet_output_image, movenet_scoreList, movenet_inference_time = movenet_pose_engine.get_result(template_image_src)
_thumbnail, posenet_output_image, posenet_scoreList, posenet_inference_time = mobilenet_pose_engine.get_result(template_image_src)
movenet_scoreList['Inference_time'] = movenet_inference_time
posenet_scoreList['Inference_time'] = posenet_inference_time
scoreList_table = pd.DataFrame(data = [movenet_scoreList, posenet_scoreList],
columns = ['LShoulder_score','RShoulder_score','LHip_score','RHip_score',
'Inference_time'],
index=['movenet','posenet'])
fig = plt.figure(figsize=(15,9))
fig.add_subplot(1, 3, 1)
plt.imshow(_thumbnail)
plt.axis("off")
plt.title("Input Image")
fig.add_subplot(1, 3, 2)
plt.imshow(movenet_output_image)
plt.axis("off")
plt.title("MoveNet")
fig.add_subplot(1, 3, 3)
plt.imshow(posenet_output_image)
plt.axis("off")
plt.title("PoseNet")
plt.show()
print()
print(scoreList_table)
print()
idx = scoreList_table.index.to_frame()
idx.insert(0, 'image', template_path)
idx.insert(1, 'model', ['movenet','posenet'])
scoreList_table = scoreList_table.set_index(pd.MultiIndex.from_frame(idx),drop=True).droplevel(2)
comparison_df = comparison_df.append(scoreList_table)
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.681930 0.584232 0.564108 0.636249
posenet 0.693424 0.361835 0.746721 0.579940
Inference_time
movenet 0.257100
posenet 0.110522
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.551457 0.685538 0.571250 0.497153
posenet 0.984250 0.821950 0.874787 0.944533
Inference_time
movenet 0.191198
posenet 0.106315
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.514819 0.482569 0.634595 0.475725
posenet 0.816953 0.787055 0.619340 0.432841
Inference_time
movenet 0.176375
posenet 0.112343
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.293247 0.280728 0.428216 0.462449
posenet 0.080133 0.064593 0.137335 0.169320
Inference_time
movenet 0.170986
posenet 0.112449
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.291344 0.409784 0.505634 0.557060
posenet 0.358750 0.380527 0.474331 0.578293
Inference_time
movenet 0.235913
posenet 0.130628
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.385434 0.443161 0.457041 0.452208
posenet 0.993103 0.882560 0.989262 0.947945
Inference_time
movenet 0.184163
posenet 0.110542
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.579216 0.608703 0.172724 0.137665
posenet 0.437934 0.132671 0.211096 0.093292
Inference_time
movenet 0.231990
posenet 0.110048
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.671805 0.560217 0.708826 0.771403
posenet 0.994041 0.837409 0.998104 0.975233
Inference_time
movenet 0.206323
posenet 0.126451
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.574028 0.661798 0.480063 0.483903
posenet 0.966646 0.948529 0.926106 0.748626
Inference_time
movenet 0.185531
posenet 0.120075
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.404683 0.323302 0.628407 0.531736
posenet 0.994282 0.981373 0.995236 0.992494
Inference_time
movenet 0.199134
posenet 0.107284
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.060427 0.090580 0.245573 0.340671
posenet 0.064809 0.060697 0.029433 0.048051
Inference_time
movenet 0.199152
posenet 0.118681
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.576427 0.593523 0.652052 0.692572
posenet 0.999428 0.971296 0.929319 0.827800
Inference_time
movenet 0.180601
posenet 0.110199
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.482163 0.407907 0.389751 0.356105
posenet 0.676122 0.718254 0.509295 0.618937
Inference_time
movenet 0.176185
posenet 0.112858
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.411386 0.387540 0.387879 0.432137
posenet 0.028315 0.037679 0.026444 0.043686
Inference_time
movenet 0.183578
posenet 0.137858
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.272308 0.308843 0.197491 0.171467
posenet 0.011118 0.017692 0.005691 0.006544
Inference_time
movenet 0.257633
posenet 0.147543
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.065165 0.159040 0.078886 0.121895
posenet 0.640908 0.661787 0.309637 0.311637
Inference_time
movenet 0.195607
posenet 0.091741
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.475422 0.576229 0.795357 0.678322
posenet 0.522550 0.907768 0.935941 0.754716
Inference_time
movenet 0.180939
posenet 0.098230
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.722023 0.518078 0.559009 0.590755
posenet 0.993478 0.997292 0.993563 0.996355
Inference_time
movenet 0.205538
posenet 0.101784
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.463635 0.651494 0.533021 0.667355
posenet 0.990964 0.883968 0.999335 0.954334
Inference_time
movenet 0.193792
posenet 0.110132
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.649309 0.577869 0.570143 0.533052
posenet 0.984572 0.941343 0.940407 0.924790
Inference_time
movenet 0.183928
posenet 0.109419
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.351813 0.285502 0.257873 0.301363
posenet 0.789852 0.866329 0.808955 0.776207
Inference_time
movenet 0.193994
posenet 0.109088
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.698905 0.492374 0.528614 0.548246
posenet 0.955995 0.962668 0.846824 0.864783
Inference_time
movenet 0.220414
posenet 0.108495
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.060427 0.090580 0.245573 0.340671
posenet 0.064809 0.060697 0.029433 0.048051
Inference_time
movenet 0.191457
posenet 0.093907
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.705921 0.629297 0.577361 0.644588
posenet 0.890897 0.707792 0.991340 0.855217
Inference_time
movenet 0.207087
posenet 0.095760
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.709016 0.488555 0.673661 0.542463
posenet 0.832334 0.937849 0.724507 0.842540
Inference_time
movenet 0.204091
posenet 0.111062
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.467571 0.432706 0.533694 0.443867
posenet 0.938513 0.992204 0.895669 0.907810
Inference_time
movenet 0.194182
posenet 0.111170
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.203492 0.159095 0.293808 0.355325
posenet 0.004774 0.005587 0.001251 0.001916
Inference_time
movenet 0.157535
posenet 0.110726
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.563815 0.644333 0.536523 0.575943
posenet 0.932518 0.951447 0.859028 0.974499
Inference_time
movenet 0.185939
posenet 0.108440
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.608532 0.695414 0.584077 0.577192
posenet 0.994645 0.998270 0.990977 0.988087
Inference_time
movenet 0.203677
posenet 0.101956
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.664439 0.450120 0.566758 0.627638
posenet 0.909101 0.721479 0.985621 0.808457
Inference_time
movenet 0.200272
posenet 0.094260
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.011379 0.027955 0.012385 0.031742
posenet 0.003143 0.001506 0.009571 0.006482
Inference_time
movenet 0.182232
posenet 0.106513
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.368593 0.407846 0.507244 0.531825
posenet 0.900541 0.994566 0.941537 0.994153
Inference_time
movenet 0.170849
posenet 0.106671
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.194734 0.153795 0.137528 0.105855
posenet 0.003191 0.005295 0.004226 0.005557
Inference_time
movenet 0.193280
posenet 0.110677
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.452736 0.561611 0.615250 0.632143
posenet 0.927438 0.873117 0.944495 0.948101
Inference_time
movenet 0.187261
posenet 0.110619
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.484825 0.600958 0.633971 0.499900
posenet 0.782442 0.947943 0.930646 0.991978
Inference_time
movenet 0.192665
posenet 0.106557
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.422470 0.647567 0.642165 0.616534
posenet 0.941262 0.934462 0.984502 0.965508
Inference_time
movenet 0.174384
posenet 0.090501
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.287767 0.225122 0.151572 0.131170
posenet 0.049259 0.039406 0.015937 0.019279
Inference_time
movenet 0.176963
posenet 0.104440
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.097631 0.090283 0.114825 0.156921
posenet 0.133590 0.085984 0.109510 0.137726
Inference_time
movenet 0.196708
posenet 0.107798
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.168734 0.167558 0.614498 0.572879
posenet 0.645109 0.945679 0.892022 0.992411
Inference_time
movenet 0.187756
posenet 0.111031
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.628801 0.674815 0.619654 0.794412
posenet 0.955991 0.907969 0.850242 0.781131
Inference_time
movenet 0.186534
posenet 0.110910
LShoulder_score RShoulder_score LHip_score RHip_score \
movenet 0.361242 0.604967 0.296043 0.329390
posenet 0.987998 0.953838 0.964929 0.861641
Inference_time
movenet 0.174212
posenet 0.108728
In [8]:
comparison_df
Out[8]:
| LShoulder_score | RShoulder_score | LHip_score | RHip_score | Inference_time | ||
|---|---|---|---|---|---|---|
| image | model | |||||
| fall_img_46.png | movenet | 0.681930 | 0.584232 | 0.564108 | 0.636249 | 0.257100 |
| posenet | 0.693424 | 0.361835 | 0.746721 | 0.579940 | 0.110522 | |
| fall_img_2.png | movenet | 0.551457 | 0.685538 | 0.571250 | 0.497153 | 0.191198 |
| posenet | 0.984250 | 0.821950 | 0.874787 | 0.944533 | 0.106315 | |
| fall_img_31.jpg | movenet | 0.514819 | 0.482569 | 0.634595 | 0.475725 | 0.176375 |
| posenet | 0.816953 | 0.787055 | 0.619340 | 0.432841 | 0.112343 | |
| fall_img_45.jpg | movenet | 0.293247 | 0.280728 | 0.428216 | 0.462449 | 0.170986 |
| posenet | 0.080133 | 0.064593 | 0.137335 | 0.169320 | 0.112449 | |
| fall_img_43.jpg | movenet | 0.291344 | 0.409784 | 0.505634 | 0.557060 | 0.235913 |
| posenet | 0.358750 | 0.380527 | 0.474331 | 0.578293 | 0.130628 | |
| fall_img_27.jpg | movenet | 0.385434 | 0.443161 | 0.457041 | 0.452208 | 0.184163 |
| posenet | 0.993103 | 0.882560 | 0.989262 | 0.947945 | 0.110542 | |
| fall_img_36.jpg | movenet | 0.579216 | 0.608703 | 0.172724 | 0.137665 | 0.231990 |
| posenet | 0.437934 | 0.132671 | 0.211096 | 0.093292 | 0.110048 | |
| fall_img_18.png | movenet | 0.671805 | 0.560217 | 0.708826 | 0.771403 | 0.206323 |
| posenet | 0.994041 | 0.837409 | 0.998104 | 0.975233 | 0.126451 | |
| fall_img_6.png | movenet | 0.574028 | 0.661798 | 0.480063 | 0.483903 | 0.185531 |
| posenet | 0.966646 | 0.948529 | 0.926106 | 0.748626 | 0.120075 | |
| fall_img_38.jpg | movenet | 0.404683 | 0.323302 | 0.628407 | 0.531736 | 0.199134 |
| posenet | 0.994282 | 0.981373 | 0.995236 | 0.992494 | 0.107284 | |
| fall_img_22.jpg | movenet | 0.060427 | 0.090580 | 0.245573 | 0.340671 | 0.199152 |
| posenet | 0.064809 | 0.060697 | 0.029433 | 0.048051 | 0.118681 | |
| fall_img_8.png | movenet | 0.576427 | 0.593523 | 0.652052 | 0.692572 | 0.180601 |
| posenet | 0.999428 | 0.971296 | 0.929319 | 0.827800 | 0.110199 | |
| fall_img_41.jpg | movenet | 0.482163 | 0.407907 | 0.389751 | 0.356105 | 0.176185 |
| posenet | 0.676122 | 0.718254 | 0.509295 | 0.618937 | 0.112858 | |
| fall_img_29.jpg | movenet | 0.411386 | 0.387540 | 0.387879 | 0.432137 | 0.183578 |
| posenet | 0.028315 | 0.037679 | 0.026444 | 0.043686 | 0.137858 | |
| fall_img_9.png | movenet | 0.272308 | 0.308843 | 0.197491 | 0.171467 | 0.257633 |
| posenet | 0.011118 | 0.017692 | 0.005691 | 0.006544 | 0.147543 | |
| fall_img_19.png | movenet | 0.065165 | 0.159040 | 0.078886 | 0.121895 | 0.195607 |
| posenet | 0.640908 | 0.661787 | 0.309637 | 0.311637 | 0.091741 | |
| fall_img_42.jpg | movenet | 0.475422 | 0.576229 | 0.795357 | 0.678322 | 0.180939 |
| posenet | 0.522550 | 0.907768 | 0.935941 | 0.754716 | 0.098230 | |
| fall_img_1.png | movenet | 0.722023 | 0.518078 | 0.559009 | 0.590755 | 0.205538 |
| posenet | 0.993478 | 0.997292 | 0.993563 | 0.996355 | 0.101784 | |
| fall_img_39.jpg | movenet | 0.463635 | 0.651494 | 0.533021 | 0.667355 | 0.193792 |
| posenet | 0.990964 | 0.883968 | 0.999335 | 0.954334 | 0.110132 | |
| fall_img_5.png | movenet | 0.649309 | 0.577869 | 0.570143 | 0.533052 | 0.183928 |
| posenet | 0.984572 | 0.941343 | 0.940407 | 0.924790 | 0.109419 | |
| fall_img_26.jpg | movenet | 0.351813 | 0.285502 | 0.257873 | 0.301363 | 0.193994 |
| posenet | 0.789852 | 0.866329 | 0.808955 | 0.776207 | 0.109088 | |
| fall_img_47.png | movenet | 0.698905 | 0.492374 | 0.528614 | 0.548246 | 0.220414 |
| posenet | 0.955995 | 0.962668 | 0.846824 | 0.864783 | 0.108495 | |
| fall_img_23.jpg | movenet | 0.060427 | 0.090580 | 0.245573 | 0.340671 | 0.191457 |
| posenet | 0.064809 | 0.060697 | 0.029433 | 0.048051 | 0.093907 | |
| fall_img_48.png | movenet | 0.705921 | 0.629297 | 0.577361 | 0.644588 | 0.207087 |
| posenet | 0.890897 | 0.707792 | 0.991340 | 0.855217 | 0.095760 | |
| fall_img_28.jpg | movenet | 0.709016 | 0.488555 | 0.673661 | 0.542463 | 0.204091 |
| posenet | 0.832334 | 0.937849 | 0.724507 | 0.842540 | 0.111062 | |
| fall_img_7.png | movenet | 0.467571 | 0.432706 | 0.533694 | 0.443867 | 0.194182 |
| posenet | 0.938513 | 0.992204 | 0.895669 | 0.907810 | 0.111170 | |
| fall_img_4.png | movenet | 0.203492 | 0.159095 | 0.293808 | 0.355325 | 0.157535 |
| posenet | 0.004774 | 0.005587 | 0.001251 | 0.001916 | 0.110726 | |
| fall_img_34.jpg | movenet | 0.563815 | 0.644333 | 0.536523 | 0.575943 | 0.185939 |
| posenet | 0.932518 | 0.951447 | 0.859028 | 0.974499 | 0.108440 | |
| fall_img_40.jpg | movenet | 0.608532 | 0.695414 | 0.584077 | 0.577192 | 0.203677 |
| posenet | 0.994645 | 0.998270 | 0.990977 | 0.988087 | 0.101956 | |
| fall_img_30.jpg | movenet | 0.664439 | 0.450120 | 0.566758 | 0.627638 | 0.200272 |
| posenet | 0.909101 | 0.721479 | 0.985621 | 0.808457 | 0.094260 | |
| background.jpg | movenet | 0.011379 | 0.027955 | 0.012385 | 0.031742 | 0.182232 |
| posenet | 0.003143 | 0.001506 | 0.009571 | 0.006482 | 0.106513 | |
| fall_img_0.png | movenet | 0.368593 | 0.407846 | 0.507244 | 0.531825 | 0.170849 |
| posenet | 0.900541 | 0.994566 | 0.941537 | 0.994153 | 0.106671 | |
| fall_img_21.jpg | movenet | 0.194734 | 0.153795 | 0.137528 | 0.105855 | 0.193280 |
| posenet | 0.003191 | 0.005295 | 0.004226 | 0.005557 | 0.110677 | |
| fall_img_35.jpg | movenet | 0.452736 | 0.561611 | 0.615250 | 0.632143 | 0.187261 |
| posenet | 0.927438 | 0.873117 | 0.944495 | 0.948101 | 0.110619 | |
| fall_img_44.jpg | movenet | 0.484825 | 0.600958 | 0.633971 | 0.499900 | 0.192665 |
| posenet | 0.782442 | 0.947943 | 0.930646 | 0.991978 | 0.106557 | |
| fall_img_37.jpg | movenet | 0.422470 | 0.647567 | 0.642165 | 0.616534 | 0.174384 |
| posenet | 0.941262 | 0.934462 | 0.984502 | 0.965508 | 0.090501 | |
| fall_img_32.jpg | movenet | 0.287767 | 0.225122 | 0.151572 | 0.131170 | 0.176963 |
| posenet | 0.049259 | 0.039406 | 0.015937 | 0.019279 | 0.104440 | |
| fall_img_24.jpg | movenet | 0.097631 | 0.090283 | 0.114825 | 0.156921 | 0.196708 |
| posenet | 0.133590 | 0.085984 | 0.109510 | 0.137726 | 0.107798 | |
| fall_img_20.png | movenet | 0.168734 | 0.167558 | 0.614498 | 0.572879 | 0.187756 |
| posenet | 0.645109 | 0.945679 | 0.892022 | 0.992411 | 0.111031 | |
| fall_img_3.png | movenet | 0.628801 | 0.674815 | 0.619654 | 0.794412 | 0.186534 |
| posenet | 0.955991 | 0.907969 | 0.850242 | 0.781131 | 0.110910 | |
| fall_img_33.jpg | movenet | 0.361242 | 0.604967 | 0.296043 | 0.329390 | 0.174212 |
| posenet | 0.987998 | 0.953838 | 0.964929 | 0.861641 | 0.108728 |
In [ ]:
Fall-Detection on Video Samples¶
In [9]:
def show_frames(frame_list, frame_score):
frame_list = frame_list[-50:]
frame_to_display = len(frame_list)
columns = 6
rows = (frame_to_display // columns) + 1
fig = plt.figure(figsize=(20, 42))
plt_cnt = 1
for i in range(0, len(frame_list)):
fig.add_subplot(rows, columns, plt_cnt)
plt.axis("off")
plt_title = ''
plt_title +='LS:'+str(frame_score[i]['LShoulder_score'])+'\n'
plt_title +='RS:'+str(frame_score[i]['RShoulder_score'])+'\n'
plt_title +='LH:'+str(frame_score[i]['LHip_score'])+'\n'
plt_title +='RH:'+str(frame_score[i]['RHip_score'])+'\n'
plt.title(plt_title)
plt.imshow(frame_list[i])
plt_cnt += 1
In [10]:
def movenet_execution_for_video(video_path):
cam = cv2.VideoCapture(video_path)
currentframe = 0
output_frame_sequences = []
output_frame_score = []
while(True):
ret,frame = cam.read()
if ret:
name = str(currentframe) + '.jpg'
currentframe += 1
template_image_src = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
template_image_src = Image.fromarray(template_image_src)
_thumbnail, movenet_output_image, movenet_scoreList, movenet_inference_time = movenet_pose_engine.get_result(template_image_src)
temp_score = sum([ 1 if e> 0.5 else 0 for e in list(movenet_scoreList.values())])
output_frame_sequences.append(movenet_output_image)
output_frame_score.append(movenet_scoreList)
else:
break
cam.release()
cv2.destroyAllWindows()
show_frames(output_frame_sequences, output_frame_score)
In [11]:
def posenet_execution_for_video(video_path):
cam = cv2.VideoCapture(video_path)
currentframe = 0
output_frame_sequences = []
output_frame_score = []
while(True):
ret,frame = cam.read()
if ret:
name = str(currentframe) + '.jpg'
currentframe += 1
template_image_src = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
template_image_src = Image.fromarray(template_image_src)
_thumbnail, posenet_output_image, posenet_scoreList, posenet_inference_time = mobilenet_pose_engine.get_result(template_image_src)
temp_score = sum([ 1 if e> 0.5 else 0 for e in list(posenet_scoreList.values())])
output_frame_sequences.append(posenet_output_image)
output_frame_score.append(posenet_scoreList)
else:
break
cam.release()
cv2.destroyAllWindows()
show_frames(output_frame_sequences, output_frame_score)
video -1¶
In [12]:
print("Posenet \n")
posenet_execution_for_video('videos/video_1.mp4')
Posenet
In [13]:
print("Movenet \n")
movenet_execution_for_video('videos/video_1.mp4')
Movenet
video - 2¶
In [14]:
print("Posenet \n")
posenet_execution_for_video('videos/video_2.mp4')
Posenet
In [15]:
print("Movenet \n")
movenet_execution_for_video('videos/video_2.mp4')
Movenet
video - 3¶
In [16]:
print("Posenet \n")
posenet_execution_for_video('videos/video_3.mp4')
Posenet
In [17]:
print("Movenet \n")
movenet_execution_for_video('videos/video_3.mp4')
Movenet
video - 4¶
In [18]:
print("Posenet \n")
posenet_execution_for_video('videos/video_4.mp4')
Posenet
In [19]:
print("Movenet \n")
movenet_execution_for_video('videos/video_4.mp4')
Movenet
video - 5¶
In [20]:
print("Posenet \n")
posenet_execution_for_video('videos/video_5.mp4')
Posenet
In [21]:
print("Movenet \n")
movenet_execution_for_video('videos/video_5.mp4')
Movenet
In [ ]:
In [ ]: