pytorch-deeplab-xception icon indicating copy to clipboard operation
pytorch-deeplab-xception copied to clipboard

inference issue

Open xiaoyuamw opened this issue 3 years ago • 5 comments

Hi,@jaemin93 I had run the inference.py provided by you successfully! But what puzzled me is that the pictures of result acquired by me are all like this: image

The real result of seg_result i want to acquire is like this:

seg_result

So, what should i do to acquire what i want? Thank you very much!

Best Regards.

xiaoyuamw avatar Apr 26 '21 12:04 xiaoyuamw

thank you. may be you can try .jpg image file. Can you try png file?

2021년 4월 26일 (월) 오후 9:43, xiaoyuamw @.***>님이 작성:

@.*** https://github.com/jaemin93 I had run the inference.py provided by you successfully! But what puzzled me is that the pictures of result acquired by me are all like this: [image: image] https://user-images.githubusercontent.com/59866090/116083516-402c0580-a6cf-11eb-95ff-2c60abd4ded1.png

The real result i want to acquire is like this: [image: seg_result] https://user-images.githubusercontent.com/59866090/116083768-9b5df800-a6cf-11eb-95c2-d477f39ca943.png

So, what should i do to acquire what i want? thank you very much!

Best Regards.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jfzhang95/pytorch-deeplab-xception/issues/204, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHDTHJPAHY4G2GIZWEIV4YLTKVNXPANCNFSM43SZ34IQ .

jaemin93 avatar Apr 26 '21 14:04 jaemin93

Thank you for your reply. The datasets i used before are the voc_2012, and the pictures included in it are .jpg image files. Meanwhile, the command i used are like those: first: python inference.py; second: python gray_to_rgb.py; third: python crop_seg_img.py.

Whether the comand lines i used are right?

Then, i found a picture of .png image file to try again, which failed again! Would you mind to give me more details about how to use the test method?

xiaoyuamw avatar Apr 26 '21 15:04 xiaoyuamw

The mask area, which is the result of segmentation, is cut from the cat's photo. So you can map the mask area from a black image to a different color.

jaemin93 avatar Apr 30 '21 07:04 jaemin93

Hi, can you share a copy of inference.py to [email protected] I couldn't find it .

XUSHITINE avatar Oct 21 '21 07:10 XUSHITINE

Hi, can you share a copy of inference.py to me?[email protected] I couldn't find it . https://github.com/jaemin93/pytorch-deeplab-xception/blob/master/inference.py

jaemin93 avatar Oct 21 '21 07:10 jaemin93