Ruben Wiersma
@rubenwiersma.nl
Postdoc at ETH Zürich, studying geometry processing and geometric deep learning. Previously PhD @TUDelft, research intern @Adobe. rubenwiersma.nl
The #SIGGRAPH Thesis Fast Forward 2026 submissions are now open! Deadline: Nov 7th 2025. Let the world know what you've been working on in your PhD! More info at research.siggraph.org/thesisff/ and check out last year's iteration youtube.com/watch?v=8esF...
October 7, 2025 at 3:06 PM
The #SIGGRAPH Thesis Fast Forward 2026 submissions are now open! Deadline: Nov 7th 2025. Let the world know what you've been working on in your PhD! More info at research.siggraph.org/thesisff/ and check out last year's iteration youtube.com/watch?v=8esF...
Mark your calendars for the #SIGGRAPH 2025 Thesis Fast Forward in-person event. Recent PhD graduates will present their thesis work and answer questions on Monday 11 August, 10:30-11:30am: s2025.conference-schedule.org/presentation...
July 30, 2025 at 7:42 AM
Mark your calendars for the #SIGGRAPH 2025 Thesis Fast Forward in-person event. Recent PhD graduates will present their thesis work and answer questions on Monday 11 August, 10:30-11:30am: s2025.conference-schedule.org/presentation...
The materials of my graduate school course at #SGP2025 on "Deep Learning on Meshes and Point Clouds" are live on rubenwiersma.nl/deeplearning. A 'map' for 3D deep learning, covering the basics and thoughts on when and where to use deep learning tools in 3D.
July 9, 2025 at 9:12 AM
The materials of my graduate school course at #SGP2025 on "Deep Learning on Meshes and Point Clouds" are live on rubenwiersma.nl/deeplearning. A 'map' for 3D deep learning, covering the basics and thoughts on when and where to use deep learning tools in 3D.
The dataset contains hyperspectral scans of nine reconstructions of historical paintings and ten historic pigments. The code provides tools to process and load this data and utilities to work with hyperspectral data (e.g., stitching scans, conversion to RGB). 2/3
June 11, 2025 at 8:50 AM
The dataset contains hyperspectral scans of nine reconstructions of historical paintings and ten historic pigments. The code provides tools to process and load this data and utilities to work with hyperspectral data (e.g., stitching scans, conversion to RGB). 2/3
A human (who cared) might have actually centered the text in the text box 😒
June 6, 2025 at 1:06 PM
A human (who cared) might have actually centered the text in the text box 😒
This gives us a practical measure of uncertainty that we use to 1. guide captures for better data, 2. share data between points, and 3. guide data-based priors in a diffusion model. 5/6
April 30, 2025 at 1:30 PM
This gives us a practical measure of uncertainty that we use to 1. guide captures for better data, 2. share data between points, and 3. guide data-based priors in a diffusion model. 5/6
Entropy can be expensive to compute. In our paper, we use frequency analysis to simplify and accelerate this computation. Our method takes 0.0005s, where it takes 6 minutes with a physically-based path tracer (Mitsuba), while still correlating strongly (0.9) 4/6
April 30, 2025 at 1:30 PM
Entropy can be expensive to compute. In our paper, we use frequency analysis to simplify and accelerate this computation. Our method takes 0.0005s, where it takes 6 minutes with a physically-based path tracer (Mitsuba), while still correlating strongly (0.9) 4/6
How do we know when this is the case? In our paper, we use **entropy** to measure uncertainty. We argue (and show in our experiments) that where there is high entropy, we are more likely to have high error. 3/6
April 30, 2025 at 1:30 PM
How do we know when this is the case? In our paper, we use **entropy** to measure uncertainty. We argue (and show in our experiments) that where there is high entropy, we are more likely to have high error. 3/6
First, an example of uncertainty. In the image on the left, it is hard to tell a matte plane from a glossy plane, while it's clear on the right. In the first case, better algorithms will not give a better estimate, because the data is limited. 2/6
April 30, 2025 at 1:30 PM
First, an example of uncertainty. In the image on the left, it is hard to tell a matte plane from a glossy plane, while it's clear on the right. In the first case, better algorithms will not give a better estimate, because the data is limited. 2/6
Recent progress in multi-view 3D object capture is stunning! But is there a limit?
In our upcoming #SIGGRAPH2025 paper "Uncertainty for SVBRDF Acquisition using Frequency Analysis," we show how to answer this question for material estimation. 1/6
In our upcoming #SIGGRAPH2025 paper "Uncertainty for SVBRDF Acquisition using Frequency Analysis," we show how to answer this question for material estimation. 1/6
April 30, 2025 at 1:30 PM
Recent progress in multi-view 3D object capture is stunning! But is there a limit?
In our upcoming #SIGGRAPH2025 paper "Uncertainty for SVBRDF Acquisition using Frequency Analysis," we show how to answer this question for material estimation. 1/6
In our upcoming #SIGGRAPH2025 paper "Uncertainty for SVBRDF Acquisition using Frequency Analysis," we show how to answer this question for material estimation. 1/6
Checking the "Submission Results" email at 3AM
April 10, 2025 at 8:28 AM
Checking the "Submission Results" email at 3AM
Found some exploratory experiments I did early in my PhD: @sniklaus.com's 3D Ken Burns effect on paintings by Pieter de Hooch. Didn't go anywhere, but still looks cool. More below.
January 28, 2025 at 2:39 PM
Found some exploratory experiments I did early in my PhD: @sniklaus.com's 3D Ken Burns effect on paintings by Pieter de Hooch. Didn't go anywhere, but still looks cool. More below.
Debugging results right until the deadline.
January 23, 2025 at 10:09 AM
Debugging results right until the deadline.
A hot glue gun was enough to link the tori into the final standing statue. If you want to read the dissertation, you can find it here :) repository.tudelft.nl/record/uuid:... 7/7
November 19, 2024 at 12:36 PM
A hot glue gun was enough to link the tori into the final standing statue. If you want to read the dissertation, you can find it here :) repository.tudelft.nl/record/uuid:... 7/7
Finally, I linked the rings together by making a small cut and gluing back the missing piece. More laser cutter video, because it's so cool 😎 6/7
November 19, 2024 at 12:35 PM
Finally, I linked the rings together by making a small cut and gluing back the missing piece. More laser cutter video, because it's so cool 😎 6/7
Etching is magical. There could have been a problem with the focal point of the laser and the curved surface, but it turned out okay in practice. 5/7
November 19, 2024 at 12:34 PM
Etching is magical. There could have been a problem with the focal point of the laser and the curved surface, but it turned out okay in practice. 5/7
The fun part was etching text on a curved surface. I used an orthographic camera to render the scene, which models how the laser cutter 'views' the object. 4/7
November 19, 2024 at 12:33 PM
The fun part was etching text on a curved surface. I used an orthographic camera to render the scene, which models how the laser cutter 'views' the object. 4/7
The first step was to get a digital copy of the wooden rings, which was as simple as measuring the radii and creating a torus in Blender. 3/7
November 19, 2024 at 12:32 PM
The first step was to get a digital copy of the wooden rings, which was as simple as measuring the radii and creating a torus in Blender. 3/7
The cover is a set of three linked tori with the title mapped on the surface. I thought about 3D printing, but figured it would be challenging to get the overhanging parts right. I eventually went with prefab wooden rings that I would etch with a laser cutter. 2/7
November 19, 2024 at 12:31 PM
The cover is a set of three linked tori with the title mapped on the surface. I thought about 3D printing, but figured it would be challenging to get the overhanging parts right. I eventually went with prefab wooden rings that I would etch with a laser cutter. 2/7