Hello sir, thanks for the tutorial, what is the fps when you were processing the video on the fly? seems slower than when you played back the processed video in the beginning
Hi thanks for video, i wonder is there a way to export with fp16?
Can we build the same model which capture real time footages not a video as a input
Input is through a external connected camera and live object detection
can we make it work the same like this
hi bro, can you do object recognition with webcam and then rotate the camera motor so that the object is in the middle of the frame (do servos and pca9685 work?)
Sir, how do you make the results from tensorrt can be used in real time like using a webcam? Can you explain step by step?
Thanks for your video, So how can i Know the memory usage of this method
Thanks for your video. Good hints for us. Can you share about the detection speed? Compare to YOLOv5 TRT on Jetson Nano. Is it faster than YOLOv5 on the same device? I have tried your tutorial and run YOLOv5 on Jetson Nano. It seems around 5-7 FPS with 640 img size input and 12-14 FPS with 320 img size.
Yolov7 is definitely faster than v5. I didn't tested with fps but will share comparison video soon.
Good day. Thanks for the video. I have the few questions for you:
1. Can I use the same method to "build engine" even if my weights are not tiny weights? I got an error when I did that until I included the t.
2. What is the reason behind the model showing bounding boxes without labels? I'm also encountering that. Please note that I'm a beginner so please trying to understand the Yolo concept.
Yes you can use the same method to build engine for any other version of yolov7 model. Make sure to use appropriate flags to build engine.
I rechecked the video and there are labels showing with bbox.
Hi, nice video. Can I use the pretrained yolov7-tiny model so that it recognizes only some of the categories? Like, just detect cats, for example?
Hi Sir.
I really appreciated your code, thank you!
Just one question... what do the four point coordinates of output bounding boxes refer to?
Are the first two the x and y of the left top corner or not?
Please I Hope you could reply
Thank you