xiaozhi2015

How many steps have you trained at stage1 and stage2, respectively?

2

@alex04072000 Does the steps in stage1 need to be the same as stage2?

How to convert other imgs through the parameters we get from one img inferring?

1

Using opencv or photoshop? Parameters include exposure, contrast, color curve, tone curve, and gamma value. @yuanming-hu

The SRN method is not stable enough? @jiangsutx

Can you share your running speed when inferring, for example, upscale a 1920 * 1080 img to 3840 * 2160? @yulunzhang Much thx.

“The version here keeps the full resolution for the best possible retrieval performance” in paper， but the input size is 10 * 3 * 224 * 224 in VGG_ILSVRC_16_pool5.prototxt. Will...

Running speed.

3

Can you share your running speed when inferring, for example, upscale a 1920 * 1080 img to 3840 * 2160? @thstkdgus35 @yulunzhang

@yaushian

@TimoBolkart And can I use my own FLAME topology model as the --template_fname ?

Artifacts when visualize the test.npy

1

 As is shown @EvelynFan Anybody came across this problem?