<div dir="ltr"><div>Hi all,</div><div><br></div><div>I'm giving my MS project defense talk today at 1pm, and I would be happy if you would come!!</div><div><br></div><div>Best,</div><div>Adam<br></div><div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">---------- Forwarded message ---------<br>From: <b class="gmail_sendername" dir="auto">Karen van Gool</b> <span dir="auto"><<a href="mailto:kvangool@ucsb.edu" target="_blank">kvangool@ucsb.edu</a>></span><br>Date: Thu, Jun 6, 2019 at 11:09 AM<br>Subject: [grads] Adam Schmidt MS Project Defense<br>To: <<a href="mailto:faculty@lists.cs.ucsb.edu" target="_blank">faculty@lists.cs.ucsb.edu</a>>, <<a href="mailto:grads@lists.cs.ucsb.edu" target="_blank">grads@lists.cs.ucsb.edu</a>>, <<a href="mailto:research@lists.cs.ucsb.edu" target="_blank">research@lists.cs.ucsb.edu</a>>, <<a href="mailto:colloquia@lists.cs.ucsb.edu" target="_blank">colloquia@lists.cs.ucsb.edu</a>><br></div><br><br><div dir="ltr"><p class="MsoNormal" style="text-align:center;margin:0in 0in 0.0001pt;font-size:12pt;font-family:Cambria,serif" align="center"><span style="font-size:24pt;font-family:"Times New Roman",serif;color:black">MS Project
Defense</span></p>
<p class="MsoNormal" style="text-align:center;margin:0in 0in 0.0001pt;font-size:12pt;font-family:Cambria,serif" align="center"><b><span style="font-size:24pt;font-family:"Times New Roman",serif;color:black">Adam Schmidt</span></b></p>
<p class="MsoNormal" style="text-align:center;margin:0in 0in 0.0001pt;font-size:12pt;font-family:Cambria,serif" align="center"><span style="font-size:24pt;font-family:"Times New Roman",serif;color:black">Wednesday,
June 12, 2019</span></p>
<p class="MsoNormal" style="text-align:center;margin:0in 0in 0.0001pt;font-size:12pt;font-family:Cambria,serif" align="center"><span style="font-size:24pt;font-family:"Times New Roman",serif;color:black">1:00 PM
– HFH 1132</span></p>
<pre style="margin:0.1pt 0in;font-size:10pt;font-family:Courier"><span style="font-size:18pt;font-family:"Times New Roman",serif;color:black"> </span></pre><pre style="margin:0.1pt 0in;font-size:10pt;font-family:Courier"><span style="font-size:18pt;font-family:"Times New Roman",serif;color:black"> </span></pre><pre style="font-size:10pt;font-family:Courier"><b><span style="font-size:16pt;font-family:"Times New Roman",serif;color:black">Committee: </span></b><span style="font-size:16pt;font-family:"Times New Roman",serif">Pradeep Sen (Chair), Lingqi Yan</span></pre>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><span style="font-size:16pt;font-family:"Times New Roman",serif"> </span></p>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><b><span style="font-size:16pt;font-family:"Times New Roman",serif">Title:</span></b><span style="font-size:16pt"> </span><span style="font-size:16pt;font-family:"Times New Roman",serif">Fast Rendered Image Denoising Using
Adaptive Blockwise Computation</span></p>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><span style="font-size:16pt;font-family:"Times New Roman",serif"> </span></p>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><span style="font-size:16pt;font-family:"Times New Roman",serif"> </span></p>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><b><span style="font-size:16pt;font-family:"Times New Roman",serif;color:black">Abstract:</span></b></p>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><span style="font-size:16pt;font-family:"Times New Roman",serif;color:black"><br>
This project looks at solutions for removing noise from rendered images in
real-time. Rendering is the process of creating an image from a model of a
scene or environment. Creating detailed renders can be an extremely costly
process, so a recent push in graphics has been to create noisy renders and then
estimate the detailed output. Given a rendered image, how can we quickly
denoise this result? Fast denoising would be useful when a game company wants
to denoise the scenes in their game in real-time, or when an animation studio
needs quick results to see how a scene will look in their final animation. This
project focuses on multiple concepts that can help machine learning models
approach denoising in real-time. The main novel concept the project introduces
is an optimization to allocate computation based on region detail. We use a
chain of multiple Convolutional Neural Networks (CNNs) to estimate a detailed
image at different levels of refinement. After each refinement a binary-mask
estimator is introduced to decide which regions have reached sufficient levels
of quality. The model then continues on more complicated image regions. In my
talk I will detail the model's results and give a brief survey on the current
state of the art.</span></p>
<p class="MsoNormal" style="font-size:12pt;font-family:Cambria,serif"><span style="font-size:16pt;font-family:"Times New Roman",serif;color:black"> <br></span></p>
<b><span style="font-size:16pt;font-family:"Times New Roman",serif;color:black">Everyone welcome!</span></b> <br clear="all"><div><br></div>-- <br><div dir="ltr" class="m_8115907033916417798m_3564424278961159586m_-582158570442432988gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div>Karen van Gool</div><div><i>she, her</i></div><div dir="ltr">Graduate Student Advisor<div>Computer Sciences,UCSB</div><div>HFH 2104</div><div>805.893.4322</div></div></div></div></div></div></div></div></div></div>
_______________________________________________<br>
grads maillist - <a href="mailto:grads@lists.cs.ucsb.edu" target="_blank">grads@lists.cs.ucsb.edu</a><br>
<a href="https://lists.cs.ucsb.edu/mailman/listinfo/grads" rel="noreferrer" target="_blank">https://lists.cs.ucsb.edu/mailman/listinfo/grads</a><br>
</div></div></div>