supernintendo
DF Pleb
First time using SAEHD
I have trained / converted using the below specs
Model name: SAEHD ==
== ==
== Current iteration: 100004 ==
== ==
==------------- Model Options --------------==
== ==
== sort_by_yaw: False ==
== random_flip: True ==
== resolution: 128 ==
== face_type: f ==
== learn_mask: True ==
== optimizer_mode: 1 ==
== archi: df ==
== ae_dims: 256 ==
== ed_ch_dims: 21 ==
== random_warp: True ==
== true_face_training: True ==
== face_style_power: 0.0 ==
== bg_style_power: 0.0 ==
== ct_mode: none ==
== clipgrad: False ==
== batch_size: 6 ==
== ==
==--------------- Running On ---------------==
== ==
== Device index: 0 ==
== Name: GeForce GTX 1060 6GB ==
== VRAM: 6.00GB ==
== ==
==============================================
Choose mode:
(0) original
(1) overlay
(2) hist-match
(3) seamless2
(4) seamless
(5) seamless-hist-match
(6) raw-rgb
(7) raw-rgb-mask
(8) raw-mask-only
(9) raw-predicted-only
Default: 4 :
4
Choose mask mode:
(1) learned
(2) dst
(3) FAN-prd
(4) FAN-dst
(5) FAN-prd*FAN-dst
(6) learned*FAN-prd*FAN-dst
?:help Default: 1 : 4
Choose erode mask modifier [-400..400] (skip:0) :
0
Choose blur mask modifier [-400..400] (skip:0) :
0
Choose motion blur power [0..100] (skip:0) :
0
Choose output face scale modifier [-50..50] (skip:0) :
0
Apply color transfer to predicted face? Choose mode ( None / rct / lct / mkl / mkl-m / idt / idt-m skip:None ) :
None
Choose sharpen mode:
(0) None
(1) box
(2) gaussian
?:help Default: 0 :
0
Choose super resolution mode:
(0) None
(1) RankSRGAN
?:help Default: 0 :
0
Choose image degrade by denoise power [0..500] (skip:0) :
0
Choose image degrade by bicubic rescale power [0..100] (skip:0) :
0
Degrade color power of final image [0..100] (skip:0) :
0
Export png with alpha channel of the mask? (y/n skip:n) :
n
This is the result
I want to improve this.Any suggestions to make it less blurry and sharper
Any comments would be very helpful
I have trained / converted using the below specs
Model name: SAEHD ==
== ==
== Current iteration: 100004 ==
== ==
==------------- Model Options --------------==
== ==
== sort_by_yaw: False ==
== random_flip: True ==
== resolution: 128 ==
== face_type: f ==
== learn_mask: True ==
== optimizer_mode: 1 ==
== archi: df ==
== ae_dims: 256 ==
== ed_ch_dims: 21 ==
== random_warp: True ==
== true_face_training: True ==
== face_style_power: 0.0 ==
== bg_style_power: 0.0 ==
== ct_mode: none ==
== clipgrad: False ==
== batch_size: 6 ==
== ==
==--------------- Running On ---------------==
== ==
== Device index: 0 ==
== Name: GeForce GTX 1060 6GB ==
== VRAM: 6.00GB ==
== ==
==============================================
Choose mode:
(0) original
(1) overlay
(2) hist-match
(3) seamless2
(4) seamless
(5) seamless-hist-match
(6) raw-rgb
(7) raw-rgb-mask
(8) raw-mask-only
(9) raw-predicted-only
Default: 4 :
4
Choose mask mode:
(1) learned
(2) dst
(3) FAN-prd
(4) FAN-dst
(5) FAN-prd*FAN-dst
(6) learned*FAN-prd*FAN-dst
?:help Default: 1 : 4
Choose erode mask modifier [-400..400] (skip:0) :
0
Choose blur mask modifier [-400..400] (skip:0) :
0
Choose motion blur power [0..100] (skip:0) :
0
Choose output face scale modifier [-50..50] (skip:0) :
0
Apply color transfer to predicted face? Choose mode ( None / rct / lct / mkl / mkl-m / idt / idt-m skip:None ) :
None
Choose sharpen mode:
(0) None
(1) box
(2) gaussian
?:help Default: 0 :
0
Choose super resolution mode:
(0) None
(1) RankSRGAN
?:help Default: 0 :
0
Choose image degrade by denoise power [0..500] (skip:0) :
0
Choose image degrade by bicubic rescale power [0..100] (skip:0) :
0
Degrade color power of final image [0..100] (skip:0) :
0
Export png with alpha channel of the mask? (y/n skip:n) :
n
This is the result
I want to improve this.Any suggestions to make it less blurry and sharper
Any comments would be very helpful