ServerAnalysisModule

설치 설명

• python 3.6

• torch 1.5

• torchvision 0.6.0

• Clone the repository
  git clone https://https://github.com/glee1228/ServerAnalysisModule.git

• Install the required packages

  pip install opencv-python
  apt-get install -y libsm6 libxext6 libxrender-dev
  pip3 install ffmpeg-python
  apt-get install -y ffmpeg

  pip install -r requirements.txt

• Weight files are uploaded to the repository(folder ./weights)

  • For Downloading the official pre-trained weights from https://github.com/leoxiaobin/deep-high-resolution-net.pytorch
    Direct links (official Drive folder, official OneDrive folder):
  • COCO w48 384x288 (more accurate, but slower) - Used as default in live_demo.py and the other scripts
    pose_hrnet_w48_384x288.pth
  • COCO w32 256x192 (less accurate, but faster)
    pose_hrnet_w32_256x192.pth
  • MPII w32 256x256 (MPII human joints)
    pose_hrnet_w32_256x256.pth

  Remember to set the parameters of SimpleHRNet accordingly.

• For multi-person support: FasterRCNN (obstacle detection)

  • Install FasterRCNN required packages (Included in above requirements.txt)

  • Weight files are uploaded to the repository(folder ./models/detectors/FasterRCNN/checkpoints)

  • Build Non Maximum Suppression and ROI Align modules (modified from facebookresearch/maskrcnn-benchmark)

  1. Install(from folder ./models/detectors/FasterRCNN)

     $ python support/setup.py develop

  2. Uninstall

     $ python support/setup.py develop --uninstall

  3. Test

     $ python test/nms/test_nms.py

     • Result

     

     • more details

     potterhsu/easy-faster-rcnn.pytorch

• (Optional) Download the COCO dataset and save it in ./datasets/COCO

• Your folders should look like:

  simple-HRNet
  ├── datasets                (datasets - for training only)
  │  └── COCO                 (COCO dataset)
  ├── losses                  (loss functions)
  ├── misc                    (misc)
  │  └── nms                  (CUDA nms module - for training only)
  ├── models                  (pytorch models)
  │  └── detectors            (people detectors)
  |    └── FasterRCNN         (FasterRCNN repository)
  │      ├── ...
  │      └── checkpoints          (FasterRCNN weights)
  │    └── yolo               (PyTorch-YOLOv3 repository)
  │      ├── ...
  │      └── weights          (YOLOv3 weights)
  ├── scripts                 (scripts)
  ├── testing                 (testing code)
  ├── training                (training code)
  └── weights                 (HRnet weights)

• If you want to run the training script on COCO scripts/train_coco.py, you have to build the nms module first.
  Please note that a linux machine with CUDA is currently required. Built it with either:

  • cd misc; make or
  • cd misc/nms; python setup_linux.py build_ext --inplace

클래스 사용방법

import cv2
from SimpleHRNet import SimpleHRNet

model = SimpleHRNet(48, 17, "./weights/pose_hrnet_w48_384x288.pth")
image = cv2.imread("image.png", cv2.IMREAD_COLOR)

joints = model.predict(image)

live demo 실행

From a connected camera:

python scripts/live_demo.py --camera_id 0

From a saved video:

python scripts/live_demo.py --filename clip1_1.avi

학습 script 실행

python scripts/train_coco.py

For help:

python scripts/train_coco.py --help

CCTV Module을 위한 Object Detection, Pose Estimation, Feature Extraction 전체 코드 실행

python infer.py --filename clip1_1.avi