brian-yu / pedestrian-cam Goto Github PK

View Code? Open in Web Editor NEW

115.0 7.0 37.0 17.6 MB

Monitoring Foot Traffic over IP Webcams with ML

Home Page: https://news.ycombinator.com/item?id=15874866

License: MIT License

Jupyter Notebook 99.81% Python 0.19%

yolo ip-webcam people-counter people-detection machine-learning computer-vision python

pedestrian-cam's Introduction

Pedestrian Cam

Counting Foot Traffic Over IP Webcams with Machine Learning

This is the repository for the project talked about in this blog post.

How to get up and running:

Ensure you have prerequisite libraries
- Install Python 3 and OpenCV python

Clone YOLO & Darknet

git clone https://github.com/pjreddie/darknet

Clone this repository into the same directory

git clone https://github.com/brian-yu/pedestrian-cam.git
mv pedestrian-cam/* .
rm -r pedestrian-cam

Download Yolo 2.0 weights

wget https://pjreddie.com/media/files/yolo.2.0.weights

Run the files
- For the webserver, run server.py and prediction.py
- Otherwise, you can explore the Jupyter notebooks.

pedestrian-cam's People

Contributors

Stargazers

Watchers

pedestrian-cam's Issues

Can you add a license?

Is it possible to add a LICENSE to this work?

Thanks

Issue running server.py

[tcp @ 0x7fa5e952a180] Connection to tcp://84.35.225.233:83 failed: Operation timed out
OpenCV: Couldn't read video stream from file "http://84.35.225.233:83/SnapshotJPEG?Resolution=640x480&amp;amp;Quality=Clarity&amp;amp;1509566566"
<class 'cv2.error'>

in prediction.py i found the line

cap = cv2.VideoCapture('http://84.35.225.233:83/SnapshotJPEG?Resolution=640x480&amp;amp;Quality=Clarity&amp;amp;1509566566')

and changed the image location to a valid url but nothing happened. How can I get this to work?

Approach to collect data

Hi,
I am new to YOLO.
I have a few question regarding your approach to collect data.

1-Why you capture screen(static image) and count people rather than analyze video instead? I am not sure whether analyze video or static image is more accurate.
2-Is your approach fast enough to analyze live video as it takes time to anaylize each image?
3-You don't set ROI, so same people will appear on multiple snapshot images. How to handle that duplicate?

Sorry, as mentioned about i am new to this Machine learning thing so don't me if my doubt sound ridiculous.

VideoCapture fails every time

Hi there,

No matter what feed I use (e.g. http://81.205.59.129:6001/mjpg/video.mjpg), cv2 video capture will fail. ret in ret, frame = cap.read() will be false, and the stacktrace says:

/Users/turbo/anaconda2/lib/python2.7/site-packages/matplotlib/axes/_axes.pyc in imshow(self, X, cmap, norm, aspect, interpolation, alpha, vmin, vmax, origin, extent, shape, filternorm, filterrad, imlim, resample, url, **kwargs)
   5192                               resample=resample, **kwargs)
   5193 
-> 5194         im.set_data(X)
   5195         im.set_alpha(alpha)
   5196         if im.get_clip_path() is None:

/Users/turbo/anaconda2/lib/python2.7/site-packages/matplotlib/image.pyc in set_data(self, A)
    598         if (self._A.dtype != np.uint8 and
    599                 not np.can_cast(self._A.dtype, float, "same_kind")):
--> 600             raise TypeError("Image data cannot be converted to float")
    601 
    602         if not (self._A.ndim == 2

TypeError: Image data cannot be converted to float

brian-yu / pedestrian-cam Goto Github PK

pedestrian-cam's Introduction

Pedestrian Cam

Counting Foot Traffic Over IP Webcams with Machine Learning

How to get up and running:

pedestrian-cam's People

Contributors

Stargazers

Watchers

Forkers

pedestrian-cam's Issues

Live CCTV

Can you add a license?

Issue running server.py

Approach to collect data

VideoCapture fails every time

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent