search

WongKinYiu / yolov7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors
GNU General Public License v3.0
13.38k stars 4.22k forks source link

How to train your segmentation data with seg in U7? What is the data set format?Thanks！！ #732

Open Ohandsomeboy opened 2 years ago

WongKinYiu commented 2 years ago

Detection (box): class x_center y_center width height Instance segmentation (polygon): class x1 y1 x2 y2 ... xn yn

Ohandsomeboy commented 2 years ago

Detection (box): class x_center y_center width height Instance segmentation (polygon): class x1 y1 x2 y2 ... xn yn

Is the instance segmentation label like this? 0 0.045180722891566265 0.6942771084337349 0.007153614457831325 0.012382864792503346 0.0416039156626506 0.6887550200803213 0.046121987951807226 0.6880856760374833 0.04593373493975904 0.6951137884872824 0.04875753012048193 0.7004685408299867 0 0.09770331325301204 0.7434738955823293 0.009036144578313253 0.009705488621151271 0.10222138554216867 0.7386211512717537 0.09920933734939759 0.7419678714859438 0.09695030120481928 0.7449799196787149 0.09318524096385543 0.7483266398929049

updatelse commented 2 years ago

Detection (box): class x_center y_center width height Instance segmentation (polygon): class x1 y1 x2 y2 ... xn yn

Is the instance segmentation label like this? 0 0.045180722891566265 0.6942771084337349 0.007153614457831325 0.012382864792503346 0.0416039156626506 0.6887550200803213 0.046121987951807226 0.6880856760374833 0.04593373493975904 0.6951137884872824 0.04875753012048193 0.7004685408299867 0 0.09770331325301204 0.7434738955823293 0.009036144578313253 0.009705488621151271 0.10222138554216867 0.7386211512717537 0.09920933734939759 0.7419678714859438 0.09695030120481928 0.7449799196787149 0.09318524096385543 0.7483266398929049

Hello, which program is used to generate x1, y1, x2, y2... xn, yn? At present, I only have JSON files marked by labelme.

JerolSOIBAM commented 2 years ago

@updatelse Use the json/coco format then convert it to Yolo v5 format. You can find the link for the code https://github.com/ultralytics/JSON2YOLO

updatelse commented 2 years ago

@updatelse Use the json/coco format then convert it to Yolo v5 format. You can find the link for the code https://github.com/ultralytics/JSON2YOLO

Thank you, i understand. Thank you very much.

laitathei commented 2 years ago

I make the tutorial starting from labeling to training or even inferencing You can watch this video https://www.youtube.com/watch?v=Rff5RhCOj30