Hsueh-Ti Derek Liu (@HTDerekLiu) profile picture

PhD student, Computer Science, University of Toronto

Toronto, ON
Joined August 2015
Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)Photo by Hsueh-Ti Derek Liu (@HTDerekLiu)
Hsueh-Ti Derek Liu retweeted
If anyone's interested in what I've been up to lately, I've been leading development of some 3D foundation models at @Roblox—like a next-gen texture generator, released today: devforum.roblox.com/t/beta-i… 4k resolution, PBR materials… and it's free. :-) Also gets some things right that are important in practice (and annoying with other tools), like generating good texture even in spots where different parts of the mesh intersect. Was happy this required not just ML, but some classic signal processing to really get right. Great work, by a great team. Has definitely been a journey of understanding how generative AI tools can be designed as just another level of abstraction for creation—rather than a “one shot” replacement for artists. Still a lot of work to do, and more to come…
11
32
5
370
15,760
Hsueh-Ti Derek Liu retweeted
Texture Generation now goes beyond color. Quality mode, now in beta, generates 4K PBR textures with color, roughness, and metalness maps right in Studio. Link below.
56
58
18
1,199
119,548
Hsueh-Ti Derek Liu retweeted
📢ProcGen3D: Learning Neural Procedural Graphs for Image-to-3D Reconstruction @xinyi092298 learns neural procedural graphs to generate high-fidelity 3D - MCTS-guided sampling maintains consistency with the input image, even from real images! Check it out: xzhang-t.github.io/project/P…
8
53
1
238
14,719
Hooooray 🙌 Massive kudos to everyone behind it. It’s more than thrilled to see a topic so close to my heart — LOD — evolve from R&D, to prototype, and to a part of SLIM, in the hands of millions of Roblox users everyday!
SLIM (Scalable Lightweight Interactive Models) is live in Studio Beta, bringing us one step closer to our vision of massive high-fidelity worlds that run on any device. SLIM automatically creates composite 3D assets in the cloud for complex models, and then auto-renders a lightweight version. devforum.roblox.com/t/studio…
1
184
Hsueh-Ti Derek Liu retweeted
Join us at the SIGGRAPH Canada Party hosted by @HTDerekLiu and me, on behalf of Canada Graphics Research! RSVP here: lu.ma/epm4b7xp #SIGGRAPH2025
4
12
2,065
Hsueh-Ti Derek Liu retweeted
Logarithmic maps are incredibly useful for algorithms on surfaces--they're local 2D coordinates centered at a given source. @yousufmsoliman and I found a better way to compute log maps w/ fast short-time heat flow in "The Affine Heat Method" presented @ SGP2025 today! 🧵
3
76
3
478
37,122
Hsueh-Ti Derek Liu retweeted
Light version of @PierreTerdiman 's Zero-Byte-BVH here: skal65535.github.io/BVH/ The idea is to order the triangles as a binary tree by groups of 6 triangles, each group defines a bounding box to test. If hit, recurse both half of the remaining triangles. @DaKangz : WebGPU!
2
26
1
194
10,802
Our #Siggraph25 work found a simple, nearly one-line change that greatly eases neural field optimization for a wide variety of existing representations. “Stochastic Preconditioning for Neural Field Optimization” w/ @merlin_ND @_AlecJacobson @nmwsharp
8
46
10
311
57,442
Hsueh-Ti Derek Liu retweeted
I've updated my blog post to walk through the remaining technical details of our Surface Winding Numbers algorithm: now the calculus of the algorithm is explained a bit more in detail. The post, paper, code, etc. is all here: nzfeng.github.io/research/WN…
My SIGGRAPH 2023 presentation of "Winding Numbers on Discrete Surfaces", authored with @MarkGillespie64 and @keenanisalive , is now on YouTube: youtu.be/QnMx3s4_4WY
7
61
4
418
37,066
Hsueh-Ti Derek Liu retweeted
After many years of giving talks, I no longer get nervous. Instead, I'm now nervous when my students give talks! Fortunately, they do an amazing job. Here's @MarkGillespie64 giving an extended talk on a new *harmonic* surface representation: youtube.com/watch?v=9h13FPuB…
2
26
200
18,194
Hsueh-Ti Derek Liu retweeted
📢📢📢 "𝐑𝐚𝐝𝐢𝐚𝐧𝐭 𝐅𝐨𝐚𝐦: Real-Time Differentiable Ray Tracing", a mesh-based 3D represention. radfoam.github.io arxiv.org/abs/2502.01157 Co-lead by my PhD students Shrisudhan Govindarajan and Daniel Rebain, and w/ @kwangmoo_yi
24
106
17
636
115,194
Hsueh-Ti Derek Liu retweeted
My superstar student @MarkGillespie64 is on the academic job market 🤩: markjgillespie.com/ Beyond lots of beautiful, deep, award-winning research in geometry processing, he's just a terrific person who we already miss having around. Twist his arm if you want him to apply!
6
1
72
15,105
Hsueh-Ti Derek Liu retweeted
Generating nice meshes in AI pipelines is hard. Our #SIGGRAPHAsia2024 paper proposes a new representation which guarantees manifold connectivity, and even supports polygonal meshes -- a big step for downstream editing and simulation. (1/N) SpaceMesh: research.nvidia.com/labs/tor… Joint work with @mli0603 @MarcTLaw @matanatzmon @FidlerSanja @james_r_lucas @JunGao33210520 @nmwsharp (1/n)
4
42
2
150
28,209
Hsueh-Ti Derek Liu retweeted
Signed distance functions (SDFs) are fundamental tools in graphics, vision, and physics simulation. But how do you get a high-quality SDF from messy, real-world input? At #SIGGRAPH2024, we introduced a simple method for turning "broken" geometry into a well-behaved SDF. <🧵>
12
152
19
944
109,660
Hsueh-Ti Derek Liu retweeted
🔍Need efficient distance queries for 2D/3D meshes? Check out FCPW – a user-friendly library in C++ and Python with GPU support! 💻 Get started here: github.com/rohan-sawhney/fcp… Also available on PyPI: pip install fcpw🐍
6
49
343
18,602
Super excited to share the new Roblox Graduate Fellowship program, a 2-year fellowship for PhD students in graphics, machine learning, communications, systems, data science, and economics. The application deadline is August 15, 2024!! gem.com/lp/roblox/ec-roblox-…
13
70
7,768
Hsueh-Ti Derek Liu retweeted
(1/n) Happy to share that our paper, "Simplicits: Mesh-Free, Geometry-Agnostic, Elastic Simulations" will be presented at #SIGGRAPH2024 research.nvidia.com/labs/tor…
7
26
4
93
11,903
Hsueh-Ti Derek Liu retweeted
Interested in learning about differential geometry and its connection to geometric computing? All material from the @CarnegieMellon course on #DiscreteDifferentialGeometry has been collected in a new webpage (videos, code, exercises, etc.). Check it out! geometry.cs.cmu.edu/ddg
7
103
5
536
34,628
Hsueh-Ti Derek Liu retweeted
#SIGGRAPH2024 We present Vertex Block Descent (VBD), the NEXT simulator.: nearly zero-compromised, fast🚀🚀, massively parallel💻💻, convergent📉📉, and unconditionally stable💎💎. Project Page: ankachan.github.io/Projects/… Code (coming soon): github.com/AnkaChan/Gaia
18
151
11
1,129
154,402
Hsueh-Ti Derek Liu retweeted
Very excited to share our work on SIGGRAPH 2024: CWF: Consolidating Weak Features in High-quality Mesh Simplification. Proj: ruixu.me/html/CWF/index.html Code: github.com/Xrvitd/CWF (some star pls) Paper:arxiv.org/abs/2404.15661 YouTube:youtu.be/QpTY-cHJJRU
7
32
5
196
20,829