smellslikeml / ActionAI

Real-Time Spatio-Temporally Localized Activity Detection by Tracking Body Keypoints
https://www.hackster.io/actionai/actionai-custom-tracking-multiperson-activity-recognition-fa5cb5
GNU General Public License v3.0
772 stars 192 forks source link

train action like walking, carrying, etc #20

Open muhk01 opened 4 years ago

muhk01 commented 4 years ago

hello thank you for sharing a great work, i just pretty new at computer vision task, if you would mind explaining more detail does this project able to do certain task like walking, carrying, standing, etc. and if you would mind to give me more guidance about how to obtain 'lstm.h5' files when doing train sequential, thank you

gohjiayi commented 4 years ago

Hope this isn't too late! First, collect video data for the actions you want to detect. Second, you'll have to edit the iva.py script. There is a section of code related to the variable WRITE2CSV; run all your video files to generate into a csv file similar to the squats_deadlifts_stand5.csv file. It's important that you only include videos with the action that you are trying to detect. Then, you can use train_sequential.py to train the lstm model.