-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
354 lines (321 loc) · 11.3 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
<script src="http://www.google.com/jsapi" type="text/javascript"></script>
<script type="text/javascript">google.load("jquery", "1.3.2");</script>
<style type="text/css">
body {
font-family: "HelveticaNeue-Light", "Helvetica Neue Light", "Helvetica Neue", Helvetica, Arial, "Lucida Grande", sans-serif;
font-weight:300;
font-size:18px;
margin-left: auto;
margin-right: auto;
width: 1100px;
}
h1 {
font-size:32px;
font-weight:300;
}
.disclaimerbox {
background-color: #eee;
border: 1px solid #eeeeee;
border-radius: 10px ;
-moz-border-radius: 10px ;
-webkit-border-radius: 10px ;
padding: 20px;
}
video.header-vid {
height: 140px;
border: 1px solid black;
border-radius: 10px ;
-moz-border-radius: 10px ;
-webkit-border-radius: 10px ;
}
img.header-img {
height: 140px;
border: 1px solid black;
border-radius: 10px ;
-moz-border-radius: 10px ;
-webkit-border-radius: 10px ;
}
img.rounded {
border: 1px solid #eeeeee;
border-radius: 10px ;
-moz-border-radius: 10px ;
-webkit-border-radius: 10px ;
}
a:link,a:visited
{
color: #1367a7;
text-decoration: none;
}
a:hover {
color: #208799;
}
td.dl-link {
height: 160px;
text-align: center;
font-size: 22px;
}
.layered-paper-big { /* modified from: http://css-tricks.com/snippets/css/layered-paper/ */
box-shadow:
0px 0px 1px 1px rgba(0,0,0,0.35), /* The top layer shadow */
5px 5px 0 0px #fff, /* The second layer */
5px 5px 1px 1px rgba(0,0,0,0.35), /* The second layer shadow */
10px 10px 0 0px #fff, /* The third layer */
10px 10px 1px 1px rgba(0,0,0,0.35), /* The third layer shadow */
15px 15px 0 0px #fff, /* The fourth layer */
15px 15px 1px 1px rgba(0,0,0,0.35), /* The fourth layer shadow */
20px 20px 0 0px #fff, /* The fifth layer */
20px 20px 1px 1px rgba(0,0,0,0.35), /* The fifth layer shadow */
25px 25px 0 0px #fff, /* The fifth layer */
25px 25px 1px 1px rgba(0,0,0,0.35); /* The fifth layer shadow */
margin-left: 10px;
margin-right: 45px;
}
.paper-big { /* modified from: http://css-tricks.com/snippets/css/layered-paper/ */
box-shadow:
0px 0px 1px 1px rgba(0,0,0,0.35); /* The top layer shadow */
margin-left: 10px;
margin-right: 45px;
}
.layered-paper { /* modified from: http://css-tricks.com/snippets/css/layered-paper/ */
box-shadow:
0px 0px 1px 1px rgba(0,0,0,0.35), /* The top layer shadow */
5px 5px 0 0px #fff, /* The second layer */
5px 5px 1px 1px rgba(0,0,0,0.35), /* The second layer shadow */
10px 10px 0 0px #fff, /* The third layer */
10px 10px 1px 1px rgba(0,0,0,0.35); /* The third layer shadow */
margin-top: 5px;
margin-left: 10px;
margin-right: 30px;
margin-bottom: 5px;
}
.vert-cent {
position: relative;
top: 50%;
transform: translateY(-50%);
}
hr
{
border: 0;
height: 1px;
background-image: linear-gradient(to right, rgba(0, 0, 0, 0), rgba(0, 0, 0, 0.75), rgba(0, 0, 0, 0));
}
</style>
<html>
<head>
<title>ASAPNet</title>
<meta property="og:image" content=""/> <!-- Facebook automatically scrapes this. Go to https://developers.facebook.com/tools/debug/ if you update and want to force Facebook to rescrape. -->
<meta property="og:title" content="Spatially-Adaptive Pixelwise Networks for Fast Image Translation" />
<meta property="og:description" content="T. Rott Shaham et al., CVPR 2021." />
<!-- Get from Google Analytics -->
<!-- Global site tag (gtag.js) - Google Analytics
<script async src=""></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-75863369-6');
</script>
</head> -->
<body>
<br>
<center>
<span style="font-size:36px">Spatially-Adaptive Pixelwise Networks for Fast Image Translation</span><br>
<span style="font-size:25px;line-height:2.0">CVPR 2021</span><br>
<table align=center width=1100px>
<table align=center width=1100px>
<tr>
<td align=center width=110px>
<center>
<span style="font-size:24px"><a href="https://tamarott.github.io/">Tamar Rott Shaham</a></span>
</center>
</td>
<td align=center width=100px>
<center>
<span style="font-size:24px"><a href="http://www.mgharbi.com">Michaël Gharbi</a></span>
</center>
</td>
<td align=center width=100px>
<center>
<span style="font-size:24px"><a href="https://richzhang.github.io/">Richard Zhang</a></span>
</center>
</td>
<td align=center width=100px>
<center>
<span style="font-size:24px"><a href="https://research.adobe.com/person/eli-shechtman/"> Eli Shechtman </a></span>
</center>
</td>
<td align=center width=100px>
<center>
<span style="font-size:24px"><a href="https://tomer.net.technion.ac.il/">Tomer Michaeli</a></span>
</center>
</td>
</tr>
</table>
<table align=center width=500px>
<tr>
<td align=center width=200px>
<center>
<span style="font-size:24px"><a href="https://arxiv.org/pdf/2012.02992.pdf">[Paper]</a></span>
</center>
</td>
<td align=center width=200px>
<center>
<span style="font-size:24px"><a href='https://github.com/tamarott/ASAPNet'>[GitHub]</a></span><br>
</center>
</td>
<td align=center width=200px>
<center>
<span style="font-size:24px"><a href="https://youtu.be/6-OfZ32CoBE">[Video]</a></span>
</center>
</td>
</tr>
</table>
</table>
</center>
<center>
<table align=center width=850px>
<tr>
<td width=260px>
<center>
<img class="round" style="width:400px" src="./resources/runtime_vs_imgsize_Mpix_v15.png"/>
</center>
</td>
</tr>
</table>
<br>
<table align=center width=850px>
<tr align=justify>
<td>
Our novel model, designed with A Spatially Adaptive Pixelwise Network (ASAPNet) enables generating high-resolution images
at significantly lower runtimes than existing methods, while maintaining high visual quality. Particularly, as seen in the plot our
model is 2-18x faster than baselines, depending on resolution.
</td>
</tr>
</table>
</center>
<hr>
<table align=center width=850px>
<center><h1>Abstract</h1></center>
<tr align=justify>
<td>
We introduce a new generator architecture, aimed at fast and efficient high-resolution image-to-image translation.
We design the generator to be an extremely lightweight function of the full-resolution image. In fact, we use <b>pixel-wise networks</b>;
that is, each pixel is processed independently of others, through a composition of simple affine transformations and nonlinearities.
We take three important steps to equip such a seemingly simple function with adequate expressivity.
First, the parameters of the pixel-wise networks are <b>spatially varying</b>, so they can represent a broader function
class than simple 1x1 convolutions. Second, these parameters are <b>predicted</b> by a fast convolutional network that processes an
aggressively low-resolution representation of the input. Third, we augment the input image by concatenating a sinusoidal encoding of
spatial coordinates, which provides an effective inductive bias for generating realistic novel high-frequency image content.
As a result, our model is up to 18x faster than state-of-the-art baselines. We achieve this speedup while generating comparable
visual quality across different image resolutions and translation domains.
</td>
</tr>
</table>
<br>
<table align=center width=850px>
<tr>
<td width=850px>
<center>
<img class="round" style="width:700px" src="./resources/res.png"/>
</center>
</td>
</tr>
</table>
<hr>
<center><h1>5-Minutes Video</h1></center>
<table align=center width=800px>
<td width=1080 colspan=7 valign=center align=center style='width:802.5pt;padding:10pt 5.4pt 20pt 5.4pt'>
<iframe width="560" height="315" src="https://www.youtube.com/embed/6-OfZ32CoBE" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</td>
</table>
<br>
<hr>
<center><h1>Implementation</h1></center>
<table align=center width=800px>
<tr><center>
<span style="font-size:24p x"> <a href='https://github.com/tamarott/ASAPNet'>[GitHub]</a>
</center>
</span>
</table>
<br>
<table align=center width=850px>
<tr>
<td width=850px>
<center>
<img class="round" style="width:700px" src="./resources/architecture_v8.png"/>
</center>
</td>
</tr>
</table>
<table align=center width=850px>
<center>
<tr align=justify>
<td>
Our model first processes the input at very low-resolution x<sub>l</sub>, to produce a tensor of weights and
biases φ<sub>p</sub>. These are upsampled back to full-resolution, where they parameterize pixelwise, spatially-varying MLPs
f<sub>p</sub> that compute the final output y from the high-resolution input x.
</tr>
</td>
</center>
</table>
<hr>
<table align=center width=600px>
<center><h1>Paper</h1></center>
<tr align=left>
<td><a href=""><img class="layered-paper-big" style="height:175px" src="./resources/ASAPNet-01.png"/></a></td>
<td><span style="font-size:14pt">T. Rott Shaham, M. Gharbi, R. Zhang, <br>E. Shechtman, T. Michaeli<br>
<b>Spatially-Adaptive Pixelwise Networks for Fast Image Translation</b><br>
CVPR 2021<br>
<!--ArXiv, 2020<br> -->
<a href="https://arxiv.org/pdf/2012.02992.pdf">[ArXiv]</a> <a href="https://openaccess.thecvf.com/content/CVPR2021/html/Shaham_Spatially-Adaptive_Pixelwise_Networks_for_Fast_Image_Translation_CVPR_2021_paper.html">[CVF]</a> <a href="./resources/SM.pdf">[Supplementals]</a> <a href="./resources/bibtex.txt">[Bibtex]</a>
<span style="font-size:4pt"><a href=""><br></a>
</span>
</td>
</tr>
</table>
<br>
<hr>
<table align=center width=900px>
<center><h1>References</h1></center>
<tr align=justify>
<td width=400px>
<left>
Xihui Liu, Guojun Yin, Jing Shao, Xiaogang Wang and Hongsheng Li,
<b>Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis,</b>
NeurIPS 2019
<br><br>
Taesung Park, Ming-Yu Liu, Ting-Chun Wang and Jun-Yan Zhu,
<b>Semantic Image Synthesis with Spatially-Adaptive Normalization,</b>
CVPR 2019
<br><br>
Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Andrew Tao, Jan Kautz, Bryan Catanzaro,
<b> High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs,</b>
CVPR 2018
</left>
<br><br>
Xiaojuan Qi, Qifeng Chen, Jiaya Jia, and Vladlen Koltun,
<b>Semi-parametric Image Synthesis,</b>
CVPR 2018
</left>
<br><br>
Qifeng Chen and Vladlen Koltun,
<b>Photographic Image Synthesis with Cascaded Refinement Networks,</b>
ICCV 2017
</left>
</td>
</tr>
</table>
<hr>
<br>
<table align=center width=900px>
<tr>
<td width=400px>
<left>
This template was originally made by <a href="http://web.mit.edu/phillipi/">Phillip Isola</a> and <a href="http://richzhang.github.io/">Richard Zhang</a> for a <a href="http://richzhang.github.io/colorization/">colorful</a> ECCV project; the code can be found <a href="https://github.com/richzhang/webpage-template">here</a>.
</left>
</td>
</tr>
</table>
<br>
</body>
</html>