-
Notifications
You must be signed in to change notification settings - Fork 0
/
Past Projects.html
260 lines (197 loc) · 14.9 KB
/
Past Projects.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
<!DOCTYPE html>
<html lang="en">
<head>
<title>Viktor Ciroski - Blogs</title>
<!--
Viktor Ciroski
https://templatemo.com/tm-538-digital-trend
-->
<meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=Edge">
<meta name="description" content="">
<meta name="keywords" content="">
<meta name="author" content="">
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1">
<link rel="stylesheet" href="css/bootstrap.min.css">
<link rel="stylesheet" href="css/font-awesome.min.css">
<link rel="stylesheet" href="css/aos.css">
<link rel="stylesheet" href="css/owl.carousel.min.css">
<link rel="stylesheet" href="css/owl.theme.default.min.css">
<!-- MAIN CSS -->
<link rel="stylesheet" href="css/templatemo-digital-trend.css">
</head>
<body>
<!-- MENU BAR -->
<nav class="navbar navbar-expand-lg">
<div class="container">
<a class="navbar-brand" href="index.html">
<i class="fa fa-car"></i>
Viktor Ciroski
</a>
<button class="navbar-toggler" type="button" data-toggle="collapse" data-target="#navbarNav" aria-controls="navbarNav" aria-expanded="false"
aria-label="Toggle navigation">
<span class="navbar-toggler-icon"></span>
</button>
<div class="collapse navbar-collapse" id="navbarNav">
<ul class="navbar-nav ml-auto">
<li class="nav-item">
<a href="index.html" class="nav-link smoothScroll">About Me</a>
</li>
<li class="nav-item">
<a href="#publications" class="nav-link smoothScroll">Publications</a>
</li>
<li class="nav-item">
<a href="#presentations" class="nav-link smoothScroll">Presentations</a>
</li>
<li class="nav-item">
<a href="#independent-projects" class="nav-link smoothScroll">Independent Projects</a>
</li>
<!--
<li class="nav-item">
<a href="Blogs.html" class="nav-link">Blogs</a>
</li>
-->
<li class="nav-item">
<a href="contact.html" class="nav-link contact">Contact</a>
</li>
</ul>
</div>
</div>
</nav>
<!-- publications -->
<section class="publications" id="publications">
<div class="container">
<div class="row">
<div class="col-lg-12 col-12 py-5 mt-5 mb-3 text-center">
<h1 class="mb-4" data-aos="fade-up">Publications</h1>
</div>
<div class="col-lg-7 col-md-7 col-12 mb-4">
<div class="Blogs-header" data-aos="fade-up" data-aos-delay="100">
<div class="Blogs-header-info">
<h4 class="Blogs-category text-info">Lane Line Detection by LiDAR Intensity Value Interpolation</h4>
<img src="images/project/LiDARLaneLines.png" class="img-fluid" alt="Blogs">
<p>
Lane marks are an important aspect for autonomous driving. Autonomous vehicles rely on lane mark information to determine a safe and legal path to drive. In this paper an approach to estimate lane lines on straight or slightly curved roads using a LiDAR unit for autonomous vehicles is presented. By comparing the difference in elevation of LiDAR channels, a drivable region is defined. The presented approach used in this paper differs from previous LiDAR lane line detection methods by reducing the drivable region from three to two dimensions exploring only the x-y trace. In addition, potential lane markings are extracted by filtering a range of intensity values as opposed to the traditional approach of comparing neighboring intensity values. Further, by calculating the standard deviation of the potential lane markings in the y-axis, the data can be further refined to specific points of interest. By applying a statistical approximation, to these points of interest, the results given show a linear approximation of the lane lines.
</p>
</div>
</div>
</div>
<div class="col-lg-5 col-md-5 col-12 mb-4">
<div class="Blogs-sidebar d-flex justify-content-center align-items-center" data-aos="fade-up" data-aos-delay="200">
<div class="Blogs-info">
<h4 class="Blogs-category text-primary">Citation</h4>
<p> Ciroski, V. and Park, J., "Lane Line Detection by LiDAR Intensity Value Interpolation," SAE Int. J. Adv. & Curr. Prac. in Mobility 2(2):939-944, 2020, <a href="https://www.sae.org/publications/technical-papers/content/2019-01-2607/"> https://doi.org/10.4271/2019-01-2607. </a> </p>
</div>
</div>
</div>
<div class="col-lg-7 col-md-7 col-12 mb-4">
<div class="Blogs-header" data-aos="fade-up" data-aos-delay="100">
<div class="Blogs-header-info">
<h4 class="Blogs-category text-info">Comparative Study on Transfer Learning for Object Classification and Detection. In: AI-enabled Technologies for Autonomous and Connected Vehicles</h4>
<img src="images/project/AI-enabled Technologies for Autonomouse and Connected Vehicles Cover.jpg" class="img-fluid" alt="Blogs" width="200" height="300" >
<p>
The recent development of deep neural learning achieved remarkable breakthroughs in object classification and detection. Deep learning has the capability of learning features automatically from data using general-purpose earning procedures. However, because deep neural networks require large amounts of data to train the parameters in the network, it is challenging to develop any object classification or detection system with a relatively small dataset. Transfer learning is an important machine learning technique that transfers the learned features in a pre-trained Convolution Neural Network (CNN) model into a new system. In this study, current state-of-the-art CNN models are reviewed in their architectures and characteristics. For the comparative study of transfer learning, the object classification and the detection system are implemented using transfer learning with six state-of-the-art CNN models. The object classification model has achieved an accuracy of 97.01% for the three-class classification task using transfer learning. Furthermore, six different Faster R-CNN architectures are implemented for object detection. The performances of the different transferred models are compared in terms of the accuracy and the deploying speed of the new model. Experiments show that transfer learning saves training time and achieves accurate performance by fine-tuning the pre-existing deep learning model.
</p>
</div>
</div>
</div>
<div class="col-lg-5 col-md-5 col-12 mb-4">
<div class="Blogs-sidebar d-flex justify-content-center align-items-center" data-aos="fade-up" data-aos-delay="200">
<div class="Blogs-info">
<h4 class="Blogs-category text-primary">Citation</h4>
<p> Park, J., Yu, W., Aryal, P., Ciroski, V. (2023). Comparative Study on Transfer Learning for Object Classification and Detection. In: Murphey, Y.L., Kolmanovsky, I., Watta, P. (eds) AI-enabled Technologies for Autonomous and Connected Vehicles. Lecture Notes in Intelligent Transportation and Infrastructure. Springer, Cham. <a href = https://doi.org/10.1007/978-3-031-06780-8_5> https://doi.org/10.1007/978-3-031-06780-8_5 </a> </p>
</div>
</div>
</div>
<div class="col-lg-7 col-md-7 col-12 mb-4">
<div class="Blogs-header" data-aos="fade-up" data-aos-delay="100">
<div class="Blogs-header-info">
<h4 class="Blogs-category text-info">An Approach for Reducing Racial Bias in Facial Monitoring Systems</h4>
<img src="images/project/Master_Thesis.png" class="img-fluid" alt="Blogs" width="400" height="300" >
<p>
The issue of racial bias in facial monitoring systems is a pressing concern in today’s technological landscape, as these systems have the potential to exacerbate existing inequities and perpetuate discrimination. To effectively reduce bias in pre-existing facial detection system we propose the addition of fine-tuning the model on a select subset underrepresented secondary features, skin tone, facial structure, or hair texture. We propose a novel solution using semi-supervised self-supervised training (S4L). Through S4L training along with the select subset of data this research shows how the quality, not quantity, of data can achieve similar results to current benchmark architectures utilizing a fraction of the training data.
</p>
</div>
</div>
</div>
<div class="col-lg-5 col-md-5 col-12 mb-4">
<div class="Blogs-sidebar d-flex justify-content-center align-items-center" data-aos="fade-up" data-aos-delay="200">
<div class="Blogs-info">
<h4 class="Blogs-category text-primary">Citation</h4>
<p> Ciroski, V. (2023). An Approach for Reducing Racial Bias in Facial Monitoring Systems. In: University of Michigan Libary: Deep Blue Documents. <a href = https://dx.doi.org/10.7302/6648> https://dx.doi.org/10.7302/6648 </a> </p>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="presentations" id="presentations">
<div class="container">
<div class="col-lg-12 col-12 py-5 mt-5 mb-3 text-center">
<h1 class="mb-4" data-aos="fade-up">Presentations</h1>
</div>
<div class="col-lg-7 col-md-7 col-12 mb-4">
<div class="Blogs-header" data-aos="fade-up" data-aos-delay="100">
<div class="Blogs-header-info">
<h4 class="Blogs-category text-info"><a href="https://carc.extension.harvard.edu/calendar_event/an-introduction-to-statistics/"> Harvard Extension School: An Introduction to Statistics </a> </h4>
<img src="images/project/HSE_Stats.jpg" class="img-fluid" alt="Blogs">
<p>
The seminar introduces the importance of understanding and evaluating statistics and their practical applications in various fields. It addresses common concerns and fears about statistics and provides strategies to overcome them. By the end of the seminar, participants will have a foundational understanding of statistics and the ability to apply statistical methods to real-world problems. differs from previous LiDAR lane line detection methods by reducing the drivable region from three to two dimensions exploring only the x-y trace. In addition, potential lane markings are extracted by filtering a range of intensity values as opposed to the traditional approach of comparing neighboring intensity values. Further, by calculating the standard deviation of the potential lane markings in the y-axis, the data can be further refined to specific points of interest. By applying a statistical approximation, to these points of interest, the results given show a linear approximation of the lane lines.
<br>
<br>
Dates: Jan. 2021, June 2021, Feb 2022, Feb. 2023
</p>
</div>
</div>
</div>
</div>
</section>
<footer class="site-footer">
<div class="container">
<div class="row">
<div class="col-lg-5 mx-lg-auto col-md-8 col-10">
<h1 class="text-white" data-aos="fade-up" data-aos-delay="100">Thanks for Visiting</h1>
</div>
<div class="col-lg-3 col-md-6 col-12" data-aos="fade-up" data-aos-delay="200">
<h4 class="my-4">Contact Info</h4>
<p class="mb-1">
<i class="fa fa-phone mr-2 footer-icon"></i>
+1 586 506 1223
</p>
<p>
<a href="#">
<i class="fa fa-envelope mr-2 footer-icon"></i>
</a>
</p>
</div>
<div class="col-lg-3 col-md-6 col-12" data-aos="fade-up" data-aos-delay="300">
<h4 class="my-4">Located</h4>
<p class="mb-1">
<i class="fa fa-home mr-2 footer-icon"></i>
Detroit, MI, USA
</p>
</div>
<div class="col-lg-4 mx-lg-auto text-center col-md-8 col-12" data-aos="fade-up" data-aos-delay="400">
<p class="copyright-text">
<br>
<a rel="nofollow noopener" href="https://templatemo.com">Design: TemplateMo</a></p>
</div>
<div class="col-lg-4 mx-lg-auto col-md-6 col-12" data-aos="fade-up" data-aos-delay="500">
<div class="col-lg-3 mx-lg-auto col-md-6 col-12" data-aos="fade-up" data-aos-delay="600">
<ul class="social-icon">
<li><a href="https://scholar.google.com/citations?hl=en&view_op=list_works&gmla=AJsN-F7m5AjZRDHU_mVlLxByU-IL-WVMjFY_BFYCDxG1fLUQXhKE9-QjGHf-G5S7SH3t6XeTiUXVIWgUAiQqfVsnz_zsbJYGcQ&user=tNnoYf8AAAAJ" class="fa fa-google"></a></li>
<li><a href="https://www.linkedin.com/in/viktor-ciroski/" class="fa fa-linkedin-square"></a></li>
</ul>
</div>
</div>
</div>
</footer>
<!-- SCRIPTS -->
<script src="js/jquery.min.js"></script>
<script src="js/bootstrap.min.js"></script>
<script src="js/aos.js"></script>
<script src="js/owl.carousel.min.js"></script>
<script src="js/custom.js"></script>
</body>
</html>