fishman2008

confusion regarding map output feature to the location of original image

1

We caculate Correspondence by this fomula: x1 = x1_map * 2 + 1, y1 = y1_map * 2 + 1, x2 = x1_map * 2 + 1 + 12, y2...

custom anchors not improving the performance

4

I use the provided scripts to generate the anchors on the wider dataset. However, the performance is not improving much compared to the original anchors. Have you observed the similar...

If just using the public CASIA-SURF dataset and following the given parameters, the ACER is about 0.1. Are the results reported in your Table using the private dataset? Anyone able...

I have trained to generate 128 by 128 images. The generated samples with fixed noise samples look good after certain iterations. However, when using random noise, the generated samples have...

-mfpu=neon-vfpv4 unrecognized

3

The library is compiled on ubuntu system, instead of Raspberry Pi. And the error shows that "-mfpu=neon-vfpv4" is unrecognized. Could this library be compiled on ubuntu with android-NDK, and then...

examples of train_file

4

Could you provide examples of the format of the train CSV file? Thanks.