Skip to content

Commit 23191eb

Browse files
kimmyjukimmyju
committed
Updated to new version of website
1 parent 241d5d4 commit 23191eb

491 files changed

Lines changed: 17815 additions & 14127 deletions

File tree

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎About.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,7 @@ permalink: /Professor/
1414
<tr>
1515
<td></td>
1616
<td></td>
17-
<td valign="top"><img src="/img/SimonWoo_new.jpg" alt="Simon Woo"/> </td>
17+
<td valign="top"><img src="/img/member_images/SimonWoo_new.jpg" alt="Simon Woo"/> </td>
1818
<td></td>
1919
<td valign="top">
2020
◼ SKKU Fellowship Professor, 2022 (<a href="https://cs.skku.edu/ko/edures/education/view/7054">News</a>)<br>

‎Dataset.md‎

Lines changed: 14 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -25,15 +25,15 @@ permalink: /Dataset/
2525
<h1 class="page-title">FakeAVCeleb Dataset</h1>
2626
<p>In FakeAVCeleb, we propose a novel Audio-Video Deepfake dataset (FakeAVCeleb) that contains not only deepfake videos but also respective synthesized lip-synced fake audios. Our FakeAVCeleb is generated using recent most popular deepfake generation methods. To generate a more realistic dataset, we selected real YouTube videos of celebrities having four racial backgrounds (Caucasian, Black, East Asian, and South Asian) to counter the racial bias issue. <p><br>
2727

28-
<p align=center><img loading="lazy" border=0 src="/img/teaser.png" width="1080"></p>
28+
<p align=center><img loading="lazy" border=0 src="/img/datasets/FakeAVCeleb/fakeceleb_nips2021.png" width="1080"></p>
2929

3030

3131
<h1 class="page-title">VFP290K dataset</h1>
3232
<p>Vision-based Fallen Person (VFP290K) dataset consists of 294,714 frames of fallen persons extracted from 178 videos from 49 backgrounds, composing 131 scenes. We empirically demonstrate the effectiveness of the features through extensive experiments comparing the performance shift based on object detection models. In addition, we evaluate our VFP290K dataset with properly divided datasets by measuring the performance of fallen person detecting systems.
3333

3434
We ranked first in the first round of the anomalous behavior recognition track of AI Grand Challenge 2020, South Korea, using our VFP290K dataset, which can further extend to other applications, such as intelligent CCTV or monitoring systems, as well. <p><br>
3535

36-
<p align=center><img loading="lazy" border=0 src="/img/VFP.JPG" width="1080"></p>
36+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/VFP.JPG" width="1080"></p>
3737

3838

3939

@@ -56,24 +56,24 @@ We ranked first in the first round of the anomalous behavior recognition track o
5656
<table>
5757
<tr>
5858
<td width="33%" valign=top>
59-
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/AGC_detection.jpg"></p>
59+
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/datasets/VFP290k/AGC_detection.jpg"></p>
6060
</td>
6161
<td width="33%" valign=top>
62-
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/AGC_detection3.jpg"></p>
62+
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/datasets/VFP290k/AGC_detection3.jpg"></p>
6363
</td>
6464
<td width="33%" valign=top>
65-
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/AGC_detection4.jpg"></p>
65+
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/datasets/VFP290k/AGC_detection4.jpg"></p>
6666
</td>
6767
</tr>
6868
<tr>
6969
<td width="33%" valign=top>
70-
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/AGC_detection1.jpg"></p>
70+
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/datasets/VFP290k/AGC_detection1.jpg"></p>
7171
</td>
7272
<td width="33%" valign=top>
73-
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/AGC_detection2.jpg"></p>
73+
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/datasets/VFP290k/AGC_detection2.jpg"></p>
7474
</td>
7575
<td width="33%" valign=top>
76-
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/AGC_detection5.jpg"></p>
76+
<p align=center><img loading="lazy" border=0 width=240 height=135 src="/img/datasets/VFP290k/AGC_detection5.jpg"></p>
7777
</td>
7878
</tr>
7979
</table>
@@ -83,24 +83,24 @@ We ranked first in the first round of the anomalous behavior recognition track o
8383
<table>
8484
<tr>
8585
<td width="33%" valign=top>
86-
<p align=center><img loading="lazy" border=0 src="/img/AGC_classification.jpg"></p>
86+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/AGC_classification.jpg"></p>
8787
</td>
8888
<td width="33%" valign=top>
89-
<p align=center><img loading="lazy" border=0 src="/img/AGC_classification1.jpg"></p>
89+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/AGC_classification1.jpg"></p>
9090
</td>
9191
<td width="33%" valign=top>
92-
<p align=center><img loading="lazy" border=0 src="/img/AGC_classification2.jpg"></p>
92+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/AGC_classification2.jpg"></p>
9393
</td>
9494
</tr>
9595
<tr>
9696
<td width="33%" valign=top>
97-
<p align=center><img loading="lazy" border=0 src="/img/AGC_classification3.jpg"></p>
97+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/AGC_classification3.jpg"></p>
9898
</td>
9999
<td width="33%" valign=top>
100-
<p align=center><img loading="lazy" border=0 src="/img/AGC_classification4.jpg"></p>
100+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/AGC_classification4.jpg"></p>
101101
</td>
102102
<td width="33%" valign=top>
103-
<p align=center><img loading="lazy" border=0 src="/img/AGC_classification5.jpg"></p>
103+
<p align=center><img loading="lazy" border=0 src="/img/datasets/VFP290k/AGC_classification5.jpg"></p>
104104
</td>
105105
</tr>
106106
</table>

‎Heroface_restoration.html‎

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
permalink: /Heroface_restoration/
44

55
---
6-
<p><a href="../Projects/">back</a></p>
6+
<p><a href="/Projects/">back</a></p>
77

88

99

@@ -12,25 +12,25 @@ <h1 class="page-title"><b>AI 기술을 활용한 6.25 전쟁영웅 사진 복원
1212
<p><br></p>
1313
<p>6.25전쟁 73주년을 맞아 대한민국의 자유를 지키기 위해 6.25 전쟁에 참전했던 용사들의 헌신을 기리며 국가보훈부와 성균관대 인공지능학과가 협업하여 참전 영웅들의 젊은 시절이 담긴 빛바랜 흑백사진을 인공지능 기술로 복원하였습니다. 노이즈 제거 (DDNM), 고해상도 복원 (GFP-GAN), iColoriT (컬러 복원), PaddleSeg (배경 합성) 등 최신 복원 기술을 사용 후 전문가의 고증을 받아 저화질 사진을 고품질의 이미지로 만들었습니다. 해당 사진들은 6.25 전쟁 73주년 전시 및 유가족에게 전달되었습니다.<p><br>
1414

15-
<p align=center><img border=0 src="/img/heroface-restoration/heroface_restoration_main.jpg" style="width:100%"></p>
15+
<p align=center><img border=0 src="/img/projects/heroface-restoration/heroface_restoration_main.jpg" style="width:100%"></p>
1616

1717

1818
<h1 class="page-title"><b>사용한 방법들</b></h1>
1919
<div style="background-color: #ABD9FE; border-radius: 16px 16px 16px 16px; margin:10px;">
2020
<p align=center style="font-size: 20px; padding-top:10px; padding-bottom:0px; margin:0px"><b>GFP-GAN</b></p>
21-
<p align=center style="padding-bottom:10px"><img src="/img/heroface-restoration/gfpgan.png" style="width:80%" ></p>
21+
<p align=center style="padding-bottom:10px"><img src="/img/projects/heroface-restoration/gfpgan.png" style="width:80%" ></p>
2222
</div>
2323
<div style="background-color: #ABD9FE; border-radius: 16px 16px 16px 16px; margin:10px; width:750">
2424
<p align=center style="font-size: 20px; padding-top:10px; padding-bottom:0px; margin:0px"><b>iColoriT</b></p>
25-
<p align=center style="padding-bottom:10px"><img src="/img/heroface-restoration/icolorit.png" style="width:80%" ></p>
25+
<p align=center style="padding-bottom:10px"><img src="/img/projects/heroface-restoration/icolorit.png" style="width:80%" ></p>
2626
</div>
2727
<div style="background-color: #ABD9FE; border-radius: 16px 16px 16px 16px; margin:10px; width:750">
2828
<p align=center style="font-size: 20px; padding-top:10px; padding-bottom:0px; margin:0px"><b>DDNM</b></p>
29-
<p align=center style="padding-bottom:10px"><img src="/img/heroface-restoration/ddnm.png" style="width:80%" ></p>
29+
<p align=center style="padding-bottom:10px"><img src="/img/projects/heroface-restoration/ddnm.png" style="width:80%" ></p>
3030
</div>
3131
<div style="background-color: #ABD9FE; border-radius: 16px 16px 16px 16px; margin:10px; width:750">
3232
<p align=center style="font-size: 20px; padding-top:10px; padding-bottom:0px; margin:0px"><b>PaddleSeg</b></p>
33-
<p align=center style="padding-bottom:10px"><img src="/img/heroface-restoration/paddleseg.png" style="width:80%" ></p>
33+
<p align=center style="padding-bottom:10px"><img src="/img/projects/heroface-restoration/paddleseg.png" style="width:80%" ></p>
3434
</div>
3535

3636

0 commit comments

Comments
 (0)