VLDB 2026 Research / reviewers in the wild / expert
Martin Taraz
dblp:292/6415
· DBLP profile ↗
6ranked-venue papers
0as first author
6since 2021 · last 2025
0000-0002-6655-9427ORCID · corroborated
Domains — the database's venue-derived domains; a paper can count in several
Human-computer interaction and ubiquitous computing · 5 · 5 since 2021Artificial intelligence and machine learning · 1 · 1 since 2021
| Year | Publication | Venue | Position |
|---|---|---|---|
| 2025 | Mallet-Based Assembly: Enabling Load-Bearing Laser-Cut ModelsabstractLaser cutting has a long tradition of building load-bearing 3D objects based on box joints and T-joints, as these joints are naturally robust against compression and shearing. Achieving robustness against tension, however, is challenging. One presumed solution is to make all joints extremely tight, to the point where they can only be assembled using a mallet. However, our survey found that making joints tight can cause models to break during assembly. In this paper, we identify the 10 underlying issues and present techniques for overcoming them: by extending parts with what we call scaffolding or by adjusting the models’ assembly order, so as to bypass states that are subject to these issues. Based on our user study and analysis of laser-cut models, scaffolding speeds up assembly for an average of 14% of the assembly operations per model, which in turn gives an average of 1.3x speed-up per model, and 70% of the models would benefit from the adjusted assembly order, that in the absence of such, would require higher assembly effort. Shohei Katakura, Chiao Fang, Mehdi Gouasmi, Lino Hellige, Yoan Tchorenev, David Bizer, Conrad Lempert, Martin Taraz, Muhammad Abdullah 0002, Patrick Baudisch |
UIST | 8 |
| 2023 | Kerfmeter: Automatic Kerf Calibration for Laser CuttingabstractWe present Kerfmeter, a hardware + software device that automatically determines how much material the laser cutter burns off, also known as kerf. Its knowledge about kerf allows Kerfmeter to make the joints of laser cut 3D models fit together with just the right tension, i.e., loose enough to allow for comfortable assembly, yet tight enough to hold parts together without glue—all this without user interaction. Kerfmeter attaches to the head of a laser cutter and works as follows: when users send a model to the laser cutter, Kerfmeter intercepts the job, injects a brief calibration routine that determines kerf, dilates the cutting plan according to this kerf, and then proceeds to fabricate the cutting plan. During the calibration routine, Kerfmeter cuts a 2cm Archimedean spiral and uses a motor to rotate it in place until it jams against the surrounding material; the angle at which the spiral jams allows Kerfmeter to infer kerf. The calibration process takes about 20s, which is >10x faster than traditional, manual kerf calibration, while also eliminating the need for expertise. In our technical evaluation, Kerfmeter produced functioning press fit joints reliably at a precision comparable to traditional manual kerf strips. Kerfmeter makes it easy to sample repeatedly; we demonstrate how this allows boosting precision past any traditional kerf strip. Shohei Katakura, Martin Taraz, Muhammad Abdullah 0002, Paul Methfessel, Lukas Rambold, Robert Kovacs, Patrick Baudisch |
CHI | 2 |
| 2023 | AirTied: Automatic Personal Fabrication of Truss StructuresabstractWe present AirTied, a device that fabricates truss structures in a fully automatic fashion. AirTied achieves this by unrolling a 20cm-wide inflatable plastic tube and tying nodes into it. AirTied creates nodes by holding onto a segment of tube, stacking additional tube segments on top of it, tying them up, and releasing the result. The resulting structures are material-efficient and light as well as sturdy, as we demonstrate by creating a 6m-tower. Unlike the prior art, AirTied requires no scaffolding and no building blocks, bringing automated truss construction into the reach of personal fabrication. Lukas Rambold, Robert Kovacs, Conrad Lempert, Muhammad Abdullah 0002, Helena Lendowski, Lukas Fritzsche, Martin Taraz, Patrick Baudisch |
UIST | 7 |
| 2022 | What's Wrong with Deep Learning in Tree Search for Combinatorial Optimization
Maximilian Böther, Otto Kißig, Martin Taraz, Sarel Cohen, Karen Seidel 0001, Tobias Friedrich 0001 |
ICLR | 3 |
| 2022 | HingeCore: Laser-Cut Foamcore for Fast AssemblyabstractWe present HingeCore, a novel type of laser-cut 3D structure made from sandwich materials, such as foamcore. The key design element behind HingeCore is what we call a finger hinge, which we produce by laser-cutting foamcore “half-way”. The primary benefit of finger hinges is that they allow for very fast assembly, as they allow models to be assembled by folding and because folded hinges stay put at the intended angle, based on the friction between fingers alone, which eliminates the need for glue or tabs. Finger hinges are also highly robust, with some 5mm foamcore models withstanding 62kg. We present HingeCoreMaker, a stand-alone software tool that automatically converts 3D models to HingeCore layouts, as well as an integration into a 3D modeling tool for laser cutting (Kyub [7]). We have used HingeCoreMaker to fabricate design objects, including speakers, lamps, and a life-size bust, as well as structural objects, such as functional furniture. In our user study, participants assembled HingeCore layouts 2.9x faster than layouts generated using the state-of-the-art for plate-based assembly (Roadkill [1]). Muhammad Abdullah 0002, Romeo Sommerfeld, Bjarne Sievers, Leonard Geier, Jonas Noack, Marcus Ding, Christoph Thieme, Laurenz Seidel, Lukas Fritzsche, Erik Langenhan, Oliver Adameck, Moritz Dzingel, Thomas Kern, Martin Taraz, Conrad Lempert, Shohei Katakura, Hany Mohsen Elhassany, Thijs Roumen, Patrick Baudisch |
UIST | 14 |
| 2021 | FastForce: Real-Time Reinforcement of Laser-Cut StructuresabstractWe present fastForce, a software tool that detects structural flaws in laser cut 3D models and fixes them by introducing additional plates into the model, thereby making models up to 52x stronger. By focusing on a specific type of structural issue, i.e., poorly connected sub-structures in closed box structures, fastForce achieves real-time performance (106x faster than finite element analysis, in the specific case of the wheelbarrow from Figure 1). This allows fastForce to fix structural issues continuously in the background, while users stay focused on editing their models and without ever becoming aware of any structural issues. Muhammad Abdullah 0002, Martin Taraz, Yannis Kommana, Shohei Katakura, Robert Kovacs, Jotaro Shigeyama, Thijs Roumen, Patrick Baudisch |
CHI | 2 |