rainofmine / Repulsion_Loss

Pytorch implementation of Repulsion Loss on RetinaNet
94 stars 21 forks source link

Label format problem #9

Open hdjsjyl opened 5 years ago

hdjsjyl commented 5 years ago

Fist, thanks for your good work. I am a little confused about the lable format. For the annotations of the pedestrian dataset, the annotation has 'ignore', 'pedestrians', 'sitting people', 'others' and so on. I try to understand the instructions of your code: $image_path/img_1.jpg x1 y1 x2 y2 person $image_path/img_1.jpg x1 y1 x2 y2 ignore $image_path/img_2.jpg . . . . .

For this format, 'person' is 'pedestrians', and 'ignore', 'sitting people', 'others' are set to 'ignore', is it correct? Any advice will be appreciated, thanks.

rainofmine commented 5 years ago

In citypersons, it is correct.