<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Projects on Ivo Gasparini</title><link>https://ivogasparini.ch/projects/</link><description>Recent content in Projects on Ivo Gasparini</description><generator>Hugo</generator><language>en</language><copyright>© 2025, Ivo Gasparini</copyright><lastBuildDate>Sun, 14 Sep 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://ivogasparini.ch/projects/index.xml" rel="self" type="application/rss+xml"/><item><title>TreeGPT: Generative Pre-trained Transformer for forestry applications with 3D point clouds</title><link>https://ivogasparini.ch/projects/treegpt/</link><pubDate>Sun, 10 Aug 2025 00:00:00 +0000</pubDate><guid>https://ivogasparini.ch/projects/treegpt/</guid><description>&lt;figure&gt;&#13;&#10; &lt;img src="https://ivogasparini.ch/posts/proj-treegpt/lidar%20forest%20plot.png" alt="LiDAR point cloud of a forest" /&gt;&#13;&#10; &lt;figcaption&gt;&lt;em&gt;Figure 1: Example of a segmented point cloud of a forest in Canton Neuchâtel used for pre-training.&lt;/em&gt;&lt;/figcaption&gt;&#13;&#10;&lt;/figure&gt;&#13;&#10;&lt;p&gt;&amp;#x1f4c4; &lt;strong&gt;&lt;a href="https://ivogasparini.ch/posts/proj-treegpt/master_thesis_gasparini.pdf" target="_blank" rel="noopener noreferrer"&gt;Master Thesis - TreeGPT&lt;/a&gt;&lt;/strong&gt;&lt;/p&gt;&#10;&lt;hr&gt;&#10;&lt;p&gt;For my master&amp;rsquo;s thesis, I tackled a challenge that has captured my interest for years: leveraging &lt;a href="https://en.wikipedia.org/wiki/Lidar" target="_blank" rel="noopener noreferrer"&gt;LiDAR&lt;/a&gt; point clouds to generate automated forest inventories.&lt;/p&gt;&#10;&lt;p&gt;I remember when I was studying forest sciences that one of the recurring problems in forest planning was the lack of accurate data about the forest area in question. How to effectively plan the allowable cut, rotation period, tree species composition, and any silvicultural intervention without an accurate overview of what we are working with? Data in this field often comes from the &lt;a href="https://www.lfi.ch/en" target="_blank" rel="noopener noreferrer"&gt;national forest inventory&lt;/a&gt;, whose resolution is inadequate when considering individual forest areas. In the past, manual sample inventories were systematically organized by public entities on a widespread basis, but nowadays costs have become prohibitive for efforts of such magnitude.&lt;/p&gt;</description></item><item><title>Pre-print: Automatic inventory of retaining walls from aerial lidar data using 3D deep learning</title><link>https://ivogasparini.ch/projects/retainfinder/</link><pubDate>Sun, 14 Sep 2025 00:00:00 +0000</pubDate><guid>https://ivogasparini.ch/projects/retainfinder/</guid><description>&lt;p&gt;The paper has been accepted! Post coming soon&amp;hellip;&lt;/p&gt;&#10;&lt;hr&gt;</description></item><item><title>Tinkering with numerical simulations: my attempt at an explicit solver for rockfall barrier analysis</title><link>https://ivogasparini.ch/projects/rockbarrier-analytica/</link><pubDate>Mon, 23 Jun 2025 00:00:00 +0000</pubDate><guid>https://ivogasparini.ch/projects/rockbarrier-analytica/</guid><description>&lt;figure&gt;&#13;&#10; &lt;img src="https://ivogasparini.ch/posts/proj-rockbarrier-analytica/rba2.jpg" alt="Rockfall barrier in RBA during impact." /&gt;&#13;&#10; &lt;figcaption&gt;&lt;em&gt;Figure 1: Rockfall barrier in RBA during impact.&lt;/em&gt;&lt;/figcaption&gt;&#13;&#10;&lt;/figure&gt;&#13;&#10;&lt;p&gt;&amp;#x1f4c4; &lt;strong&gt;&lt;a href="https://ivogasparini.ch/posts/proj-rockbarrier-analytica/RockBarrierAnalytica - report.pdf" target="_blank" rel="noopener noreferrer"&gt;RockBarrierAnalytica - Report&lt;/a&gt;&lt;/strong&gt;&lt;/p&gt;&#10;&lt;hr&gt;&#10;&lt;p&gt;As part of my work at &lt;a href="https://www.bfh.ch/en/about-bfh/people/fdn7gei7if66/" target="_blank" rel="noopener noreferrer"&gt;BFH&lt;/a&gt;, we &lt;a href="https://www.bfh.ch/en/research/all-our-consulting-services/assessment-authority-rockfall-protection-nets/" target="_blank" rel="noopener noreferrer"&gt;inspect rockfall barriers sold in Switzerland&lt;/a&gt; to check their eligibility for federal subsidies and issue quality certifications. Rockfall barriers are mechanically complex systems due to the extreme dynamics of the processes involved and can behave unpredictably. To better understand the behavior of these systems, numerical simulations can be a powerful tool. However, existing commercial software solutions often come with limitations, such as high costs or lack of modeling flexibility.&lt;/p&gt;</description></item><item><title>How I built this site</title><link>https://ivogasparini.ch/projects/build/</link><pubDate>Sun, 15 Jun 2025 00:00:00 +0000</pubDate><guid>https://ivogasparini.ch/projects/build/</guid><description>&lt;p&gt;I built this site to learn, share and connect. To share and connect, a blog format made sense. To learn, I wanted to build it from the ground up, or at least as much as reasonable. I also wanted to keep costs low, ideally below 5 CHF/month, which is about 1/3 of the price of most site builders but without any of the limitations.&lt;/p&gt;&#10;&lt;h2 id="tech-stack"&gt;Tech stack&lt;/h2&gt;&#10;&lt;p&gt;For the frontend, I went with Hugo, a static site generator written in Go. I feel like static sites are underrated these days. Raw HTML from the server is fast and efficient. I picked the beautiful &lt;a href="https://github.com/nanxiaobei/hugo-paper" target="_blank" rel="noopener noreferrer"&gt;hugo-paper&lt;/a&gt; theme and modded it a bit (contact form, multi-language support, sticky header, and a bunch of other tweaks).&lt;/p&gt;</description></item></channel></rss>