Does yolov5 have some way to detect single type object?

[abczww]

❔Question

I am using yolov5 to detect objects by webcam, yolov5 could detect so many type objects, but I don't care about most type of them, I just need single type, like: person. I think if yolov5 just detect single type object it should work faster. But I didn't find a way to impleent this requirement. I have to add some extra codes to filter the result to get all persons which means the project became slower theoretically.

In a nutshell, I want to yolov model have many types and could detect all types objects, and it should could detect some specified type objects or single type object either. e.g.: I trained many types in model including: 'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train', 'truck', 'boat', 'traffic light' But I just want to detect 'person', 'bicycle', Or just 'person',

Additional context

[github-actions[bot]]

Hello @abczww, thank you for your interest in our work! Please visit our Custom Training Tutorial to get started, and see our Jupyter Notebook , Docker Image, and Google Cloud Quickstart Guide for example environments.

If this is a bug report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom model or data training question, please note Ultralytics does not provide free personal support. As a leader in vision ML and AI, we do offer professional consulting, from simple expert advice up to delivery of fully customized, end-to-end production solutions for our clients, such as:

• Cloud-based AI systems operating on hundreds of HD video streams in realtime.
• Edge AI integrated into custom iOS and Android apps for realtime 30 FPS video inference.
• Custom data training, hyperparameter evolution, and model exportation to any destination.

For more information please visit https://www.ultralytics.com.

[glenn-jocher]

@abczww detect.py can filter detections by class: https://github.com/ultralytics/yolov5/blob/1ddf692bf89ac47d5e2dc8333cd0f4f145bb7f61/detect.py#L160

To detect only people in images: python detect.py --classes 0

[jomdolalas]

@abczww Hi! Would like to know how you were able to use your webcam as the input? The code !python detect.py --source 0 does not work for me. I am using a laptop with a webcam installed. Do I need to use an external one?

[Frank1126lin]

@abczww Hi! Would like to know how you were able to use your webcam as the input? The code !python detect.py --source 0 does not work for me. I am using a laptop with a webcam installed. Do I need to use an external one?

@jomdolalas just use --source 0 or --source 1 etc is ok.

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[TuyenSD]

i want to detect just 1 biggest object, so what shoud i do, thks for listening

[glenn-jocher]

@TuyenSD simply select the largest detection.

[mintu07ruet]

@glenn-jocher , I was using yolov5+deep tracker (https://github.com/mikel-brostrom/Yolov5_DeepSort_Pytorch) to track the following objects- 0= Person 1= Bicycle 3= motorcycle 7=truck 17=Horse 36= Skateboard Although car is tracking properly but frame is not hold continuously for bicycle and pedestrian. Any suggestions to improve the tracking! Here is the output video https://drive.google.com/file/d/16ZfNxG8p0dEpZZJ9I2vZ__jGJ4IT8w2I/view?usp=sharing Thanks!