search

leoxiaobin / deep-high-resolution-net.pytorch

The project is an official implementation of our CVPR2019 paper "Deep High-Resolution Representation Learning for Human Pose Estimation"
https://jingdongwang2017.github.io/Projects/HRNet/PoseEstimation.html
MIT License
4.31k stars 908 forks source link

Same input but different output #214

Open acmff22 opened 3 years ago

acmff22 commented 3 years ago

I edit the test.py to detect a single image keypoints. I run the code several times, for the same input image, the outputs are different each time. I wonder what causes this? Thanks!

hebangwen commented 3 years ago

For the neural network, its output is not certain. You can output your metricsto determine whether it's good or bad