-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
221 lines (205 loc) · 11.2 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
<!DOCTYPE HTML>
<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<!-- Hi, Jon Here. Please DELETE the two <script> tags below if you use this HTML, otherwise my analytics will track your page -->
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-7580334-2"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-7580334-2');
</script>
<title>Hamed Damirchi</title>
<meta name="author" content="Jon Barron">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" type="text/css" href="stylesheet.css">
</head>
<body>
<table style="width:100%;max-width:800px;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr style="padding:0px">
<td style="padding:0px">
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr style="padding:0px">
<td style="padding:2.5%;width:63%;vertical-align:middle">
<p style="text-align:center">
<name>Hamed Damirchi</name>
</p>
</p>
<p style="text-align:center">
<a href="mailto:[email protected]">Email</a>  / 
<a href="data/HamedDamirchi-CV.pdf">CV</a>  / 
<a href="https://www.linkedin.com/in/hamed-damirchi-69801a197/">LinkedIn</a>
</p>
</td>
</tr>
</tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<heading>Research</heading>
<p>
I'm interested in computer vision and visual perception.
My research path aims towards designing strcutures and imposing constraints that would allow for inferring the relationships between the entities in the physical world and their environment.
</p>
</td>
</tr>
</tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src='images/UVO.png' width="160">
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<br>
<a href="https://arxiv.org/abs/2107.00366">
<papertitle>A Consistency-Based Loss for Deep Odometry Through Uncertainty Propagation</papertitle>
</a>
<br>
<strong>Hamed Damirchi</strong>,
<a href="http://notjustatheory.github.io/">Rooholla Khorrambakht</a>,
<a href="https://aras.kntu.ac.ir/taghirad/">Hamid D. Taghirad</a>,
<a href="https://uwaterloo.ca/electrical-computer-engineering/profile/bmoshiri">Behzad Moshiri</a>
<br>
ICRA, 2023</a>
<br>
<a href="https://arxiv.org/abs/2107.00366">arXiv</a>
<p></p>
<p>We introduce a consistency-based loss function for deep odometry by compounding the estimated SE(3) pose and uncertainties. The compounded terms are then used in a negative log-likelihood objective function where the precision matrices weighting the global loss term are based on the integrated uncertainty.</p>
</td>
</tr>
<tr onmouseout="nerfw_stop()" onmouseover="nerfw_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src='images/preintegration.png' width="160">
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://arxiv.org/pdf/2007.02929.pdf">
<papertitle>Preintegrated IMU Features For Efficient Deep Inertial Odometry</papertitle>
</a>
<br>
<a href="http://notjustatheory.github.io/">Rooholla Khorrambakht</a>,
<strong>Hamed Damirchi</strong>,
<a href="https://aras.kntu.ac.ir/taghirad/">Hamid D. Taghirad</a>
<br>
<em>TinyML</em>, 2022
<br>
<a href="https://cms.tinyml.org/wp-content/uploads/talks2022/2007.02929.pdf">TinyML</a><br>
<a href="https://arxiv.org/pdf/2007.02929.pdf">arXiv</a>
<p></p>
<p>A computationally efficient inertial representation for deep inertial odometry is proposed by replacing the raw IMU data in deep Inertial models with preintegrated features to improve the model’s efficiency.</p>
</td>
</tr>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src='images/capsnet.png' width="160">
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<br>
<a href="https://arxiv.org/abs/2007.03063">
<papertitle>ARC-Net: Activity Recognition Through Capsules</papertitle>
</a>
<br>
<strong>Hamed Damirchi</strong>,
<a href="http://notjustatheory.github.io/">Rooholla Khorrambakht</a>,
<a href="https://aras.kntu.ac.ir/taghirad/">Hamid D. Taghirad</a>
<br>
<a href="https://www.icmla-conference.org/icmla20/index.html"><em>ICMLA</em>, 2020</a>
<br>
<a href="https://arxiv.org/abs/2007.03063">arXiv</a>
<p></p>
<p>A single CNN encoder is used to extract the features from multiple IMU sensors. These features are used as entities that represent patterns in the input and through dynamic routing, activity of the subject is derived in that window of inputs. By using this architecture, various benefits such as interpretability and robustnetss to noise are gained.</p>
</td>
</tr>
<tr onmouseout="nerfie_stop()" onmouseover="nerfie_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<div class="one">
<div class="two" id='nerfie_image'><video width=160 muted autoplay loop>
<source src="data/avo-dynamic.mp4" type="video/mp4">
Your browser does not support the video tag.
</video></div>
<img src='data/avo_still.png' width="160">
</div>
<script type="text/javascript">
function nerfie_start() {
document.getElementById('nerfie_image').style.opacity = "1";
}
function nerfie_stop() {
document.getElementById('nerfie_image').style.opacity = "0";
}
nerfie_stop()
</script>
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<a href="https://arxiv.org/pdf/2011.08634v1.pdf">
<papertitle>Exploring Self-Attention for Visual Odometry</papertitle>
</a>
<br>
<strong>Hamed Damirchi</strong>,
<a href="http://notjustatheory.github.io/">Rooholla Khorrambakht</a>,
<a href="https://aras.kntu.ac.ir/taghirad/">Hamid D. Taghirad</a>
<br>
<em>arXiv</em>, 2020
<br>
<a href="https://arxiv.org/pdf/2011.08634v1.pdf">arXiv</a>
<p></p>
<p>We show that common CNN-LSTM architectures fail to extract adequate motion features in the presense of artifacts in the input. Then, by using a single-head spatial self-attention mechanism and without augmentations, we show that the network is able to reject dynamic objects and focus on the background of the scene for better odometry.
</p>
</td>
</tr>
<tr onmouseout="ff_stop()" onmouseover="ff_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src='images/arasiref.png' width="160">
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<br>
<a href="https://ieeexplore.ieee.org/document/9071852">
<papertitle>ARAS-IREF: An Open-Source Low-Cost Framework for Pose Estimation</papertitle>
</a>
<br>
<strong>Hamed Damirchi</strong>,
<a href="http://notjustatheory.github.io/">Rooholla Khorrambakht</a>,
<a href="https://aras.kntu.ac.ir/taghirad/">Hamid D. Taghirad</a>
<br>
<em>ICROM</em>, 2019
<br>
<a href="https://ieeexplore.ieee.org/document/9071852">IEEE</a>
<p></p>
<p>In order to be able to collect necessary data for data driven algorithms, we needed a pose estimation framework that would be accurate while providing ground truth at a fairly high frequency. ARAS-IREF is composed of a camera alongside a linux enabled dev board that allows for pose estimation at 100+Hz.</p>
</td>
</tr>
<tr onmouseout="ff_stop()" onmouseover="ff_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src='images/calib.png' width="160">
</td>
<td style="padding:20px;width:75%;vertical-align:middle">
<br>
<a href="https://ieeexplore.ieee.org/document/9071903">
<papertitle>A Calibration Framework for Deployable Cable Driven Parallel Robots with Flexible Cables</papertitle>
</a>
<br>
<a href="http://notjustatheory.github.io/">Rooholla Khorrambakht</a>,
<strong>Hamed Damirchi</strong>,
<a href="https://aras.kntu.ac.ir/taghirad/">Hamid D. Taghirad</a>
<br>
<em>ICROM</em>, 2019
<br>
<a href="https://ieeexplore.ieee.org/document/9071903">IEEE</a>
<p></p>
<p>An effective framework for calibrating the kinematic parameters of suspended cable driven parallel robots with no requirements for expensive tools and measurement devices is proposed where this algorithm utilizes the existing force sensors in the cable robot to nominate the best set of data for calibration.</p>
</td>
</tr>
</tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr>
<td style="padding:0px">
<br>
<p style="text-align:right;font-size:small;">
The template is from <a href="https://jonbarron.info/">John Barron</a>'s <a href="https://github.com/jonbarron/website">rep</a>.
</p>
</td>
</tr>
</tbody></table>
</td>
</tr>
</table>
</body>
</html>