ambakick / person-detection-and-tracking Goto Github PK

Hello, I found a performance issue in the definition of testResizeToRangeWithPadToMaxDimensionReturnsCorrectShapes, core/preprocessor_test.py, tf.shape(out_image) will be created repeatedly during the program executing, resulting in reduced efficiency. I think it should be created before the loop.

Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.

I thank you for releasing a great project.
I want to add the value of the x axis. Where should I modify?

Hey, how can I run this project dude. Can you tell me?

help me please!
i need make detection person multiple cameras

how to do it?

Hello

I am new to all this thing, I have seen the app and it is very good.
I want to freeze the app convert to exe from python but getting Error " No module named 'tensorflow_core.python'"
can you guide me to this

Thanks in advance

Hi,thanks for your project. I have installed all the tensorflow environments,I tried to run the Person_det_track.py,but it didn't work.So I want to know how to run this project?Because I didn't find a demo.py.Thanks. @ambakick

Hi,
you said that SSD achieves 75.1% mAP, outperforming a comparable state of the art Faster R-CNN model. how about comparison with yolov3 on time and accuracy?

and how can use your ssd-inception-v2.pb to detect my custom videos?

Hi, I wanted to as like is it possible to assign a unique ID to a person, also the id remains the same when that same person is out of frame and comes back rather assigning different tracking id?

Thank you

After installing tensorflow and opencv on windows, what should I do to make it work? I mean how to use it?

How to test on my own video?

Hi Guys! First of all Thank you for such a good implementation of People detection and tracking and making it opensource.
I am trying to run your code but have some errors in the initial loop cases, as for begining the tracking has some problems.
I tried some changes but am scared that I might mess up the code, can you please guide me a bit or change in the main Person-det-track code??

Traceback (most recent call last):
File "Person_det_track.py", line 224, in
new_img = pipeline(img)
File "Person_det_track.py", line 114, in pipeline
= assign_detections_to_trackers(x_box, z_box, iou_thrd = 0.3)
File "Person_det_track.py", line 68, in assign_detections_to_trackers
if (IOU_mat[m[0],m[1]]<iou_thrd):
IndexError: index 1 is out of bounds for axis 0 with size 1

Hi,first thanks very much.I want to try to run on my own video,but it didn't work,so what should I do to test on my video?Should i write a demo.py?

How many human can this program detect?
How can I increase the maximum detectable number of human?