Simon Weber

computer vision | @UniofOxford, @ELLISforEurope member | phd @TU_Muenchen
also on LinkedIn:

[Oscar Breiner @oscar_br_tum]

Really exciting work on Functional Attention! This feels especially useful for learning operators on PDEs, physical simulations, and geometric data where the underlying structure matters.

[Maolin Gao @GaoMaolin]

We introduce a novel functional attention mechanism, with the insight inspired by geometry. Collaboration of #TUM, #Oxford and #UTAustin.

[Simon Weber @SimWeberTUM]

@oscar_br_tum @Jiefang_Xiao @GaoMaolin @stevenygd yeah, great collab between @UniofOxford, @tumcvg and @UTAustin!

[Simon Weber @SimWeberTUM]

What if attention wasn't about matching tokens, but operating in function space?

Glad to share our #ICML2026 paper:
📄 Functional Attention: From Pairwise Affinities to Functional Correspondences

w/ @Jiefang_Xiao @GaoMaolin @stevenygd Daniel Cremers
📄 https://t.co/rhn9NtwrBm

[Simon Weber @SimWeberTUM]

Why a functional view of attention?
✅ Linear-time complexity
✅ Better generalization across resolutions and discretizations
✅ State-of-the-art results on operator learning, PDE modeling, point cloud segmentation, and OOD generalization
Code: https://t.co/vsHepaZgDD

[TUM Computer Vision Group @tumcvg]

@tumcvg goes #ICCV2025 in Hawaii! 🛫🌋

We are very proud of our students who will present five papers (+ 1 workshop) during the conference!

In particular, check out Back-on-track, which is an award candidate. (Congrats @wrchen530!)

[Simon Weber @SimWeberTUM]

Looking forward to presenting our paper "Finsler multi-dimensional scaling" at #CVPR2025 on Sunday 10:30am, Poster 462!
We investigate a largely uncharted research direction in computer vision, aka Finsler manifolds...

[Lu Sang @Lu_Sang__]

Excited to be attending CVPR 2025 this week in Nashville!
I’ll be presenting our recent work:
“4Deform: Neural Surface Deformation for Robust Shape Interpolation”
📍 Poster session: [13th June 4pm-6pm poster #111]
#CVPR2025 #computervision

[Simon Weber @SimWeberTUM]

Check out our paper📝 https://t.co/v4KmYHsOlr and our video📹 https://t.co/NDPfRU1EMX

[Simon Weber @SimWeberTUM]

Looking forward to presenting our paper "Finsler multi-dimensional scaling" at #CVPR2025 on Sunday 10:30am, Poster 462!
We investigate a largely uncharted research direction in computer vision, aka Finsler manifolds...

[Simon Weber @SimWeberTUM]

After #CVPR2024, we further investigate beyond-Euclidean computer vision and extend the fundamental task of dimensionality reduction to the largely uncharted Finsler manifolds, with core applications in data analysis and visualisation!

[Simon Weber @SimWeberTUM]

Very glad to announce that our "Finsler Multi-Dimensional Scaling" paper, accepted at #CVPR2025, is now on Arxiv! https://t.co/v4KmYHsOlr

[TUM Computer Vision Group @tumcvg]

We are thrilled that our group has twelve papers accepted at #CVPR2025! 🚀
Congratulations to all of our students for this great achievement! 🎉

For more details, check out: https://t.co/pTTuMlMUdp

[TUM Computer Vision Group @tumcvg]

We are thrilled that our group has twelve papers accepted at #CVPR2025! 🚀
Congratulations to all of our students for this great achievement! 🎉

For more details, check out: https://t.co/pTTuMlMUdp

[Linus Härenstam-Nielsen @LinusHNielsen]

We are presenting DiffCD tomorrow morning at #ECCV2024!

Come by poster 214 to discuss how to fit neural SDFs to point clouds, from a geometric perspective

[Simon Weber @SimWeberTUM]

@Rafael_L_Spring Hey! Iterative re-weighting is possible. However robust cost functions as in "Robust Bundle Adjustment Revisited" (Zach, 2014) is quite difficult (afaik) and still an open challenge for the variable projection algorithm

[Simon Weber @SimWeberTUM]

Very glad to announce that our paper "Power Variable Projection for Initialization-Free Large-Scale Bundle Adjustment" has been accepted at #ECCV2024!
Check out https://t.co/eoP38Ye5DD

[Simon Weber @SimWeberTUM]

Looking forward to present "PoVar" at #ECCV2024 tomorrow at 4:30pm, Poster 097 !
Check out our paper 📝 : https://t.co/eoP38Ye5DD
our code 💻 : https://t.co/MauFu01LoC and our video 📹: https://t.co/G1BjKm3Nh1

[Simon Weber @SimWeberTUM]

Very glad to announce that our paper "Power Variable Projection for Initialization-Free Large-Scale Bundle Adjustment" has been accepted at #ECCV2024!
Check out https://t.co/eoP38Ye5DD

[Yuesong Shen @ysngshn]

Feel free to drop by our poster stand tomorrow afternoon (Thursday 13:30-15:00, hall c 4-9, poster number #1402): https://t.co/sBZkJdzk3x 😁

[Simon Weber @SimWeberTUM]

As a follow-up paper to our CVPR'23 "Power Bundle Adjustment" https://t.co/PBP6GKUrPp , we study now the BA problem without initialization and design an efficient solver based on the largely unexplored Variable Projection algorithm

[Simon Weber @SimWeberTUM]

Check out our project page 🖥️ https://t.co/fkWYzTigEK and our video 🎬 https://t.co/lVtOKDFlEm !

[Simon Weber @SimWeberTUM]

Looking forward to present our poster "Flattening the Parent Bias: Hierarchical Semantic Segmentation in the Poincaré Ball" at #CVPR2024 in Arch 4A-E, Friday 21st, 5pm!

[Pascal Mettes @PascalMettes]

1. "Flattening the Parent Bias: Hierarchical Semantic Segmentation in the Poincaré Ball" by @SimWeberTUM et al.

Standard segmentation suffers from a bias to the parent class. Hyperbolic space is not biased due to more uniform separation of class margins, no hierarchies needed!