no mask with the output image

Question

no mask with the output image

imenselmi opened this issue 2 years ago · comments

when i run the demo file i got photo without mask or bounding box (the original photo ) how can i fix this problem

imen_selmi · Answer 1 · Mon Aug 15 2022 23:54:51 GMT+0800 (China Standard Time)

runcell(15, '/home/iovision/Desktop/imen/yolov7_instance_seg/yolo7_instance_seg/train_yolov7.py')
Install mish-cuda to speed up training and inference. More importantly, replace the naive Mish with MishCuda will give a ~1.5G memory saving during training.
[08/15 16:51:11 detectron2]: Arguments: Namespace(confidence_threshold=0.21, config_file='output/clothes_yolomask_inst/config.yaml', input='test_images/a.jpg', nms_threshold=0.6, opts=['MODEL.WEIGHTS', 'output/clothes_yolomask_inst/model_final.pth'], output=None, wandb_entity=None, wandb_project=None, webcam=False)
16:51:11 08.15 INFO yolomask.py:86]: YOLO.ANCHORS: [[142, 110], [192, 243], [459, 401], [36, 75], [76, 55], [72, 146], [12, 16], [19, 36], [40, 28]]
16:51:11 08.15 INFO yolomask.py:90]: backboneshape: [64, 128, 256, 512], size_divisibility: 32
[[142, 110], [192, 243], [459, 401], [36, 75], [76, 55], [72, 146], [12, 16], [19, 36], [40, 28]]
[08/15 16:51:12 fvcore.common.checkpoint]: [Checkpointer] Loading from output/clothes_yolomask_inst/model_final.pth ...
416 416 600
confidence thresh: 0.21
0%| | 0/1 [00:00<?, ?it/s]image after transform: (600, 416, 3)
cost: 0.1843249797821045, fps: 5.425200649321252
QObject::moveToThread: Current thread (0x5605625c3c20) is not the object's thread (0x560563347de0).
Cannot move to target thread (0x5605625c3c20)

QObject::moveToThread: Current thread (0x5605625c3c20) is not the object's thread (0x560563347de0).
Cannot move to target thread (0x5605625c3c20)

MagicSource · Answer 2 · Tue Aug 16 2022 10:14:12 GMT+0800 (China Standard Time)

maybe could try check mAP see if the model converged.