final project for the Computer Vision course @UniPD. A.Y. 2021-2022
##Compilation on Linux
mkdir build
cd build
cmake ..
make
##Run the demo
./hand_detect
runs the demo on the whole test set, showing the results. Please remember to press a key after the final detection or segmentation result is shown! by default, the segmentation process also shows the intermediate steps (meanshift, bounding boxes and snakes).
##Just one image To run the whole process on one single image:
./hand_detect -d 1 -s 1 -image path_to_image -bb path_to_bound_box
##Detailed commands
-d 0
Detect and don't show the resulting image-d 1
Detect and show the resulting image-s 0
Segment and don't show the intermediate steps-s 1
Segment and show the intermediate steps-image path
Specify the path to the image you want to process. If this parameter is not provided, the computation will be on the whole test set.-bb path
Specify the folder in which the bounding boxes file will be written-h
Show a help message and exit