Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

About "seg_label": #120

Open
xu19971109 opened this issue May 23, 2022 · 1 comment
Open

About "seg_label": #120

xu19971109 opened this issue May 23, 2022 · 1 comment

Comments

@xu19971109
Copy link

The new annotation file provided by BDD100K provides the mask map directly, while YOLOP author handles json file——gen_bDD_seglabel.py with different filling values from the mask map generated by the official.I don't have the old annotations, how can I haddle the AutoDriveDataset.py with mask?

if self.cfg.num_seg_class == 3:
_,seg0 = cv2.threshold(seg_label[:,:,0],128,255,cv2.THRESH_BINARY) # 二值分割
_,seg1 = cv2.threshold(seg_label[:,:,1],1,255,cv2.THRESH_BINARY)
_,seg2 = cv2.threshold(seg_label[:,:,2],1,255,cv2.THRESH_BINARY)
else:
_,seg1 = cv2.threshold(seg_label,1,255,cv2.THRESH_BINARY)
_,seg2 = cv2.threshold(seg_label,1,255,cv2.THRESH_BINARY_INV)
_,lane1 = cv2.threshold(lane_label,1,255,cv2.THRESH_BINARY)
_,lane2 = cv2.threshold(lane_label,1,255,cv2.THRESH_BINARY_INV)

用官方给的mask图做这个分类的时候好像有点问题,作者原来的json处理方式其中一些字段又和现在的json字段不同,处理不成作者需要的那种mask结果,有没有朋友知道怎么处理这一块啊

@LuthraBhomik
Copy link

for seg_label and lane_label, YOLOP code is picking up these from the mask maps already provided. Do you want to create mask maps again ? It is not picking up the masks from the json, only the path is being used from json.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants