search

zhangboshen / A2J

Code for paper "A2J: Anchor-to-Joint Regression Network for 3D Articulated Pose Estimation from a Single Depth Image". ICCV2019
MIT License
287 stars 46 forks source link

Your work seems to be good,As a output I'm output I'm getting this when i run itop_side.py and itop_top.py,Is that correct? #19

Open Naveennavik opened 4 years ago

Naveennavik commented 4 years ago

output are:-python itopside.py 406it [00:27, 15.02it/s] ('Accuracy:', 0) ('joint', 0, 'Head', ', accuracy: ', 0) ('joint', 1, 'Neck', ', accuracy: ', 0) ('joint', 2, 'RShoulder', ', accuracy: ', 0) ('joint', 3, 'LShoulder', ', accuracy: ', 0) ('joint', 4, 'RElbow', ', accuracy: ', 0) ('joint', 5, 'LElbow', ', accuracy: ', 0) ('joint', 6, 'RHand', ', accuracy: ', 0) ('joint', 7, 'LHand', ', accuracy: ', 0) ('joint', 8, 'Torso', ', accuracy: ', 0) ('joint', 9, 'RHip', ', accuracy: ', 0) ('joint', 10, 'LHip', ', accuracy: ', 0) ('joint', 11, 'RKnee', ', accuracy: ', 0) ('joint', 12, 'LKnee', ', accuracy: ', 0) ('joint', 13, 'RFoot', ', accuracy: ', 0) ('joint', 14, 'LFoot', ', accuracy: ', 0) I couldnt get the output of images in matplot graph? MAy know if anyone isssues

Naveennavik commented 4 years ago

Which code to run to get ITOP_1.png as an output?

zhangboshen commented 4 years ago

Oh, this result seems very wrong. My guess is that there may be something wrong with your data.

johnsk95 commented 4 years ago

I'm getting the same results. @Naveennavik did you find what the problem was?

StrongMan988 commented 2 years ago

@zhangboshen Hello, my senior. I‘m trying to recurrence your nice work, and then I have had the same problem like bro Naveennavik's. I used the datasets which is you provided ITOP Body Pose Dataset in file named 'README.md'. I got it and run the file named 'data_preprosess.py' . At least, I run the file 'itop_side.py' which is subfloder of src folder ,then, I got this wrong result. So ,can you teach me what I did wrong?