Raphaël
Hi,hassony. Thanks for reply. I get more confused after reading your reply.These are 3 questions: 1. As I can see from [Codalab Not aligned result](https://competitions.codalab.org/competitions/22485#results), you report the evaluation result...
> or with wflw and 300w datasets in training and testing, can you tell me how to calculation the correct annotation center_w and center_h? Thank you very much! Have you...
> Have you sloved the problem? Have you sloved the problem?
Have you sloved the problem?
> It is a common practice to compensate for the quantization effect. Since the landmark is encoded in a heatmap of lower resolution (generally 1/4th) than the original image, when...
> any directio In my own understanding, the `+0.5` is for the interpolation to decrease the final scaled error. The 1/4 offeset is considering the heatmap , but do you...
> any directio @kuldeepbrd1 In my own understanding, the `+0.5` is for the interpolation to decrease the final scaled error. The 1/4 offeset is considering the heatmap , but do...
Training on my own dataset, I have a very interesting result here with the same so-called post-processing . In the evaluation process, the `NME' first drop and then increase at...
> Hi, Jim, @ChiHangChen Thanks for your valuable issue. It seems that you are expert in heat map encoding and decoding from the issue. Could you please help me explain...
> nsorflow codes run well. Does it have you solved the problem?