정준섭(Joonsup Jeong)
정준섭(Joonsup Jeong)
I don't think training has been done properly. Can you show the detected image?
I didn't understand the question.
You can use [inference.py](https://github.com/penny4860/Kitti-road-semantic-segmentation/blob/master/inference.py). Please refer to readme.md for datasets and pretrained models.
You can use two load buttons at the same time. For example, one annotation is marked with a red box and another annotation is marked with a blue.
You need to download svhn images. Refer https://github.com/penny4860/svhn-voc-annotation-format#1-download-svhn-images