-
Notifications
You must be signed in to change notification settings - Fork 1.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Action Recognition via IP Camera Stream #666
base: master
Are you sure you want to change the base?
Action Recognition via IP Camera Stream #666
Conversation
…y frame over the specified IP and port. * Stream on Pillow, OpenCV etc. With 3rd libraries, the visual process goes through certain stages. * Frames received from the stream are passed through the action recognition algorithm with predict_frames(). * Output of action recognition and frames read from ip camera are displayed in a widget on the screen.
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for this contribution! I left a couple of comments in the code. More importantly however this would introduce a lot of duplicate code (since only the footage input changes, but not most of the other logic).
Also this is introducing a new dependency of this repo on OpenCV (see comment I left in code) which has its own challenges.
I believe the best way forward, given that most is duplicate code, could be to not add new functionality in the nodebook but rather to add a quick appendix with a few sentences explaining how one would implement IP camera streaming themselves.
"import os\n", | ||
"from time import sleep, time\n", | ||
"from threading import Thread\n", | ||
"from IPython.display import Video\n", | ||
"\n", | ||
"# Third party tools\n", | ||
"import cv2\n", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This would add a new dependency on opencv to this repository. We made the decision to avoid such (at least no hard requirement) since OpenCV can lead to deployment problems.
"\n", | ||
" w_text.value = \"checking 'is_playing_ip_camera'\"\n", | ||
" \n", | ||
" vidcap = cv2.VideoCapture('http://admin:password@192.168.1.97:8082/')\n", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should avoid hard-coded url.
"cell_type": "markdown", | ||
"metadata": {}, | ||
"source": [ | ||
"Download the video to our data folder" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This cell is at the wrong place, should be one above/below.
Description
In the
computervision-recipes/scenarios/action_recognition/00_webcam.ipynb
study, the feature of performing the action recognition process for the instant stream provided over the information of an ip camera, similar to the action recognition process on the frame provided by the webcam.Related Issues
Resolved #665
Checklist:
staging
and notmaster
.staging
and notmaster
.staging
.