li-haoran

Results 3 comments of li-haoran

Thanks for your sharing result. Mine seems going to bad result. i also realize this human scale issue, so i use the beam_width is the ratio of the width and...

I trained 100 epochs, actually almost 50~60 epochs convergence. The loss ![plot?](https://github.com/li-haoran/mxnet-cmu_pose/blob/master/loss/open_pose.png) is quiet small. the heatmap is perfect, but the paf is not good. i think my ratio 0.2...

after i fix some bugs, the heatmap converge seems good. i post some samples and loss fig, both show the results is good.