Sinisa Stekovic

71 posts

Sinisa Stekovic

Sinisa Stekovic

@stekovic_sinisa

Trying to understand scenes in 3D @tugraz https://t.co/5RVzEvhJbw

Graz, Austria Katılım Aralık 2010
229 Takip Edilen103 Takipçiler
Sinisa Stekovic retweetledi
Arif Ahmad
Arif Ahmad@arif_ahmad_py·
We need more senior researchers camping out at their posters like this. Managed to catch 10 minutes of Alyosha turning @anand_bhattad’s poster into a pop-up mini lecture. Extra spark after he spotted @jathushan. Other folks in the audience: @HaoLi81 @konpatp @GurushaJuneja.
English
25
146
1.4K
202.1K
Sinisa Stekovic
Sinisa Stekovic@stekovic_sinisa·
At @NeurIPSConf 📢 We show how to transfer appearance between assets of significant geometric difference, e.g., between a giraffe 🦒 and a chair 🪑. Our method guides pre-trained rectified flow models for robust, geometry-aware 3D appearance transfer between objects.
Sayan Deb Sarkar@debsarkar_sayan

A little late but just in time post for #NeurIPS ⏰ 📢 At @NeurIPSConf 2025, we will be presenting "GuideFlow3D: Optimization-Guided Rectified Flow For Appearance Transfer" 🎉 We introduce a training-free appearance transfer pipeline robust to strong geometric variations between 3D objects. 📝 arXiv: arxiv.org/abs/2510.16136 A thread 🧵 1/

English
0
0
2
120
Sinisa Stekovic
Sinisa Stekovic@stekovic_sinisa·
Do you need more data for your supervised models? Our SCANnotate++ provides high-quality CAD model annotations for ScanNet++. Models trained on our automatic annotations achieve good performance for point cloud completion and single-view CAD model retrieval and alignment.
Yuchen Rao@YuchenRao

🚀"Leveraging Automatic CAD Annotations for Supervised Learning in 3D Scene Understanding" We provide high-quality CAD model and pose annotations for the ScanNet++ v1 dataset created with SCANnotate++ method. 🔗 Project Page: stefan-ainetter.github.io/SCANnotatepp/ #3DVision #ComputerVision

English
0
2
6
309
Sinisa Stekovic retweetledi
#CVPR2026
#CVPR2026@CVPR·
Behind every great conference is a team of dedicated reviewers. Congratulations to this year’s #CVPR2025 Outstanding Reviewers! #all-outstanding-reviewer" target="_blank" rel="nofollow noopener">cvpr.thecvf.com/Conferences/20…
English
5
25
189
105.1K
Sinisa Stekovic retweetledi
Imagine-ENPC
Imagine-ENPC@ImagineEnpc·
Looking forward to #CVPR2025! We will present the following papers:
English
1
4
11
883
Sinisa Stekovic
Sinisa Stekovic@stekovic_sinisa·
PyTorchGeoNodes is now on GitHub: github.com/vevenom/pytorc… PyTorchGeoNodes is a differentiable module for reconstructing 3D objects from images using interpretable shape programs. Collaboration with: Stefan Ainetter Mattia D’Urso Friedrich Fraundorfer @VincentLepetit2
Sinisa Stekovic tweet media
VincentLepetit@VincentLepetit2

Procedural modeling is a popular tool for many applications, but extremely challenging to recover from images. With PyTorchGeoNodes, we introduce a PyTorch differentiable module for rendering procedural modeling programs. vevenom.github.io/pytorchgeonode… @stekovic_sinisa

English
1
2
6
663
Sinisa Stekovic retweetledi
VincentLepetit
VincentLepetit@VincentLepetit2·
We also show how to use our PyTorchGeoNodes module for reconstructing 3D objects from images in the form of interpretable modelling programs. Work done with @stekovic_sinisa, Stefan Ainetter, Mattia D'Urso, Friedrich Fraundorfer vevenom.github.io/pytorchgeonode…
VincentLepetit tweet media
English
0
3
14
1.3K
Sinisa Stekovic retweetledi
VincentLepetit
VincentLepetit@VincentLepetit2·
Procedural modeling is a popular tool for many applications, but extremely challenging to recover from images. With PyTorchGeoNodes, we introduce a PyTorch differentiable module for rendering procedural modeling programs. vevenom.github.io/pytorchgeonode… @stekovic_sinisa
VincentLepetit tweet media
English
1
6
33
2.1K
Sinisa Stekovic
Sinisa Stekovic@stekovic_sinisa·
Just randomly ran into this post at medium.com: @mayankprashar295/reproduction-of-paper-montefloor-extending-mcts-for-reconstructing-accurate-large-scale-floor-d5433e1ccc3e" target="_blank" rel="nofollow noopener">medium.com/@mayankprashar… I am super happy to read that students at @tudelft did a great job reimplementing our MonteFloor 🫡😊They even did some additional analysis to highlight potential issues of our approach 😅
Sinisa Stekovic@stekovic_sinisa

I am happy to share some insights on our MonteFloor method, to be presented at #ICCV2021 next week (oral). We reconstruct complex floor plans from input point clouds. Paper: arxiv.org/abs/2103.11161 Project: tugraz.at/index.php?id=5…

English
0
0
6
336
Sinisa Stekovic
Sinisa Stekovic@stekovic_sinisa·
Example annotations for a ScanNet scene from our SCANnotateDataset:
English
1
0
0
117

Keşfet

@anand_bhattad @jathushan @HaoLi81 @konpatp @GurushaJuneja @NeurIPSConf @YuchenRao @ducha_aiki