I

MACHINE LEARNING
STYLE TRANSFER OF GEOMETRY FIELD ONTO XIONG’AN CITY


USE OF DIFFERENT FIELD TYPES ONTO THE SAME MAP

01


[ANXIN-L]

Different weights & number of iteration

ANXIN-L_FIELD5 [2048-10-10-500]
ANXIN-L_FIELD6.1 [2048-10-5-250]




[ANXIN-M]

Different weights

ANXIN-M_FIELD1 [2048-10-10-250]
ANXIN-M_FIELD6.3 [2048-5-5-250]



[ANXIN-S]

Different image size & number of iteration

ANXIN-S_FIELD6.4 [2048-5-5-500]
ANXIN-S_FIELD6.2-zoom [1536-5-5-200]



DIFFERENT FIELD SCALE

02


[RONGCHENG-L]

Different weights

RONGCHENG-L_Field5 (L) [2048-10-10-500]
RONGCHENG-L_Field6.4(M) [2048-5-5-500]
RONGCHENG-L_Field1-zoom(S) [2048-5-5-500]



[RONGCHENG-M]

Different weights

RONGCHENG-M_Field6.2 (L) [2048-10-10-500]
RONGCHENG-M_Field6.1(M) [2048-5-5-500]
RONGCHENG-M_Field6.3(S) [2048-15-15-500]



[RONGCHENG-S]

Different weights & number of iterations

RONGCHENG-S_Field5 (L) [2048-10-10-500]
RONGCHENG-S_Field6.1(M) [2048-5-5-250]
RONGCHENG-S_Field6.1-zoom(S) [2048-5-5-500]



Same Fields Different Parameters

03


[XIONG-L]

Same style data, different image size, weights & number of iteration

XIONG-L_Field5 [2048-10-10-500]
XIONG-L_Field5 [1536-5-5-250]
……………………………………………………….
XIONG-L_Field6.2 [2048-5-5-500]
XIONG-L_Field6.2 [1536-10-10-250]



[XIONG-M]

Different parameters with same style data

XIONG-M_Field1 [2048-10-5-500]
XIONG-M_Field1 [1536-5-10-250]



[XIONG-M2]

XIONG-M2_Field6.3 [2048-10-10-500]
XIONG-M2_Field6.3 [1536-5-5-250]



[XIONG-S]

XIONG-S_Field5 [2048-10-10-500]
XIONG-S_Field5 [1536-5-5-250]



As image size increases, the clearer the image is.
As the number of iteration increase, the more dense the image is with more repetition of geometry.


DIFFERENT FIELD SCALE

04


[SHAOCHEDIAN-L]

SHAOCHEDIAN-L_FIELD 5 (L) [2048-10-10-500]
SHAOCHEDIAN-L_FIELD 1 (M) [2048-5-5-500]
SHAOCHEDIAN-L_FIELD 6.1 (S) [2048-5-5-500]



II

MACHINE LEARNING
STYLE TRANSFER OF REFERENCE CITIES ONTO XIONG’AN


[Anxin-M_SG-M]

1024-10-10-500


[ANXIN-L_TOKYOCHIYODA-L]

2048-5-5-500


[Baiyangdian-L_SG-S]

1024-5-5-500


[Intersection1_TokyoChiyoda-L]

1024-5-5-500


[Rongcheng-M_Barcelona-L]

2048-5-5-500


[Shaochedian-L_HK-S]

1024-5-5-500


[Xiong-L_TokyoChiyoda-L]

1024-5-5-200


[Xiong-S_Canberra-S]

2048-5-5-500


III

MACHINE LEARNING
STYLE TRANSFER OF GEOMETRY FIELD ONTO REFERENCE CITIES


[ABERDEEN-L]

DIFFERENT FIELD WITH DIFFERENT SCALES


[ABERDEEN-S]

DIFFERENT FIELD WITH DIFFERENT SCALES


[HAMBURG-M]

DIFFERENT FIELD WITH DIFFERENT SCALES


[HAMBURG-S]

DIFFERENT FIELD WITH DIFFERENT SCALES


[SG-L]

DIFFERENT FIELD WITH DIFFERENT SCALES


[SG-M]

DIFFERENT FIELD WITH DIFFERENT SCALES


[SG-S]

DIFFERENT FIELD WITH DIFFERENT SCALES


[VENICE-S]

DIFFERENT FIELD WITH DIFFERENT SCALES


IV

MACHINE LEARNING
STYLE TRANSFER OF REFERENCE CITIES ONTO XIONG’AN


[Anxin-L+ABD-S-Field6.4 2048-10-10-50]


[Anxin-M_SG-S-Field5 2048-10-10-500]


[Baiyangdian-S_SG-S-Field5-zoom 1536-5-5-200]


[Xiong-M2_ABD-S-Field6.4 1536-5-5-200]


[Xiong-S_SG-S-Field5-zoom 1536-5-5-200]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s