Qixing Huang
@qixing_huang
Followers
4K
Following
4K
Media
85
Statuses
1K
I am an associate professor at UT Austin. I do research at the intersection of computer graphics, computer vision, and machine learning.
Austin, TX
Joined May 2016
https://t.co/0dckFDgx6r Very happy to get an NSF grant on injecting physical, topological, and geometric priors into 3D generative models. Usually this is not a big deal for a tenured professor, but I am excited because this is something that I have been working on recently.
9
1
101
Yunpeng @Byp215Bai is looking for a summer internship next year, please reach out to him!!
🚀 Do world models need explicit 3D? Our answer: if you’re using Transformers, introducing 3D into DiT’s positional encoding is a natural choice. 📄 Paper: https://t.co/wO6gH2aYJI 🌐 HomePage: https://t.co/jSBZDQHonN 💻 Code: https://t.co/G8OwIWVNdP
1
3
49
Bai et al., "Positional Encoding Field" Make your RoPE encoding 3D by including a z axis, then manipulate your image by simply manipulating your positional encoding in 3D --> novel view synthesis. Neat idea.
8
56
484
🚀 Do world models need explicit 3D? Our answer: if you’re using Transformers, introducing 3D into DiT’s positional encoding is a natural choice. 📄 Paper: https://t.co/wO6gH2aYJI 🌐 HomePage: https://t.co/jSBZDQHonN 💻 Code: https://t.co/G8OwIWVNdP
3
51
405
Introducing Yunpeng’s recent work positional encoding field (PE-Field), a simple 3D-aware NVS method that fine-tunes a diffusion prior with a novel positional encoding scheme. Project page: https://t.co/2kTZRNdXOO Arxiv paper: https://t.co/ldvivVNOIF
3
24
173
Jiaxin is presenting his paper Jigsaw++ which leverages a diffusion prior for fractured object reassembly.
0
1
10
Hanwen’s @hanwenjiang1 paper, RayZer, just received the Best Student Paper Honorable Mention at #ICCV2025! 🏆 He is presenting it in the Oral session this afternoon. Incredibly proud of his work and this well-deserved recognition! Webpage: https://t.co/JmwVeWKk0o
4
7
134
Thrilled to present 2 papers at #ICCV2025 ! 🎉 💃🏻 HUMOTO: THE dataset for HOI 🧩 Jigsaw++: generative model for reassembly I'm graduating in 2026 and actively exploring opportunities! If you're hiring or know of relevant positions, DMs open! See you in Honolulu! 🌺
2
7
75
UT mathematician Maggie Miller was named a 2025 Packard Fellow for Science & Engineering! 🤘 Miller is a topologist who studies geometric properties in 3, 4 & 5 dimensions. #PackardFellows #Topology #GeometricTopology #UTAustin #TexasScience @UTAustin
https://t.co/PN2knEdcDX
cns.utexas.edu
University of Texas at Austin mathematician and topological researcher Maggie Miller has won a 2025 Packard Fellowship.
0
1
5
I am attending ICCV from Saturday (AC workshop) to the final day. Two of my current students Jiaxin Lu and Yunpeng Bai will be there. My former student hanwen jiang @hanwenjiang1 will present an oral paper, which was nominated for a best paper award. I wish him good luck. I was
Six papers were accepted by ICCV 25, including five papers from three very strong students: @hanwenjiang1, @jacinth_lu, and Yunpeng Bai ( https://t.co/XrSerFGuIW). There will be big news on recent success in faculty recruitment at UT in graphics/vision. Stay tuned!
0
2
28
Six papers were accepted by ICCV 25, including five papers from three very strong students: @hanwenjiang1, @jacinth_lu, and Yunpeng Bai ( https://t.co/XrSerFGuIW). There will be big news on recent success in faculty recruitment at UT in graphics/vision. Stay tuned!
0
5
92
🚀 The full HUMOTO dataset is now licensed to leading universities around the world — including Stanford, MIT, KAIST, CMU, UCLA, Tel Aviv University, University of Tübingen & MPI, Queen Mary University of London, and many more across the US, Europe, and Asia 🌍 Since launch,
6
26
201
Speaker schedule is finalized! Unfortunately, Joao and Vincent couldn’t join this time, but we’re excited to host Bill Freeman and @qixing_huang! 📅 Oct 19 (Sun) | 📍 Room 323A See you soon! #ICCV2025
Can scaling data and models alone solve computer vision? 🤔 Join us at the SP4V Workshop at #ICCV2025 in Hawaii to explore this question! 🎤 Speakers: @danfei_xu, @joaocarreira, @jiajunwu_cs, Kristen Grauman, @sainingxie, @vincesitzmann 🔗 https://t.co/pH1Qjc1Kr2
2
7
55
Joining The University of Texas at Austin and HIRING Excited to share I will join @UTAustin as an Asst. Professor, starting January 2026. We are expanding, hiring grad students and a postdoc. Job description is in the attached files. Pls share. Website: https://t.co/xznchK5owO
13
66
345
UTCS continues to hire tenured/tenure-track faculties during the 2025-2026 cycle. We have 6+ positions across a wide-range of areas. We hired seven tenure-track faculties during the 2024-2025 cycle. You can find the application site at
0
0
8
Check out our recent work on fast physical property prediction for multi-material objects from images, led by our amazing student @LongLeRobot ! We’ve also released our annotations and data generation pipeline ( https://t.co/S4rvzdYq4S) for 3D objects with physical material labels
github.com
Feed-forward model for predicting 3D physics with 3DGS + NeRF - vlongle/pixie
NeRFs & 3DGS look great but are static. Veo/Genie show how exciting dynamic worlds can be. What if we brought that dynamism into explicit 3D? ✨ Meet Pixie: a feed-forward “Physics from Pixels” model — simulatable, fast, and generalizable 👇🧵 🌐 https://t.co/LQlobSgI14
0
8
48
One more week to submit to our #ICCV2025 Workshop on 3D Modeling, Reconstruction, and Generation in the Wild! Besides standard 3D topics, broader 4D and video-related directions also count! The papers are **non-archival** Feel free to submit your accepted papers! (1/2)
1
21
78
Sending our congratulations to @PeterStone_TX, who has just been named Chair of the Department of Computer Science at @UTAustin! @UTexasResearch Peter Stone to Lead UT Department of Computer Science as Chair, Emphasizing Continued Growth and Community https://t.co/7tU7THl5ru
0
9
75