For Anaconda environments:
conda create -n OccS_env python=3.6
conda activate OccS_env
conda install -c conda-forge tensorflow-gpu=1.14
conda install -c conda-forge opencv
conda install -c conda-forge shapely
conda install -c conda-forge matplotlib
conda install -c conda-forge xlsxwriter
- Model: -m, --model, required=True → Path to object detection model (inference graph)
- Labels: -l, --labels, required=True → Path to labels file (labelmap)
- Input: -i, --input, required=True → Path to input image file
- Output: -o, --output, default="results/output.jpg" → Path to optional output image file
- Threshold: -t, --threshold, default=0.8 → Minimum probability to filter detections
- Calibration: -c, --calibration, action="store_true" → Option for un-distort input image
- Resize: -r, --resize, → Resize input image in format "1,1" (comma separated)
- Camera Height: -H, --camera_height → z-coordinate for camera positioning
- People Height: -p, --people_height → z-coordinate for people height
- Angle: -a, --angle → positioning angle in degrees