search

Qidian213 / deep_sort_yolov3

Real-time Multi-person tracker using YOLO v3 and deep_sort with tensorflow
GNU General Public License v3.0
1.65k stars 593 forks source link

How do I track more than one class? #96

Closed luiscosio closed 5 years ago

luiscosio commented 5 years ago

Hello!

How can I use this to track more than one class at the same time? (Ex. Persons and cars)

Regards!

kazuya-n commented 5 years ago

just remove this line https://github.com/Qidian213/deep_sort_yolov3/blob/fbc993948e9d61b2e1e0367ab0edf3e7d08022fd/yolo.py#L100

but deep sort pre-trained model mars-small128.pb is trained to track only person

luiscosio commented 5 years ago

Thank you @kazuya-n, any idea how can I use different color bounding boxes for each class is tracking and different counters for each class?

So for example, I can track cars and pedestrians, each one of those categories with different colors and different counts.

luiscosio commented 5 years ago

@PassantReyad To track persons and cars you need to edit line 101 for this:

if predicted_class != 'person' and predicted_class != 'car':
                continue
liujs1016 commented 4 years ago

Did you solve it? Can you tell me how to solve it?