iiinova — where your digital self keeps evolving

Vasileios Belagiannis

@bazilas.bsky.social

Professor at FAU. Past: Univ. Magdeburg, Ulm Univ., VGG Oxford University, TU München + Industry. personal account. www.machinelearning.tf.fau.de

Posts Replies Media

Vasileios Belagiannis @bazilas.bsky.social · 29/03/2025

Dive into the details and code: 📖 lnkd.in/d4FjwFY7 (paper) 🖨️ lnkd.in/dkdabrZ9 (pre-print) 💾 lnkd.in/dfnUk8-C (code) #FAU #DFG #MachineLearning #ComputerVision

lnkd.in

This link will take you to a page that’s not on LinkedIn

000

Vasileios Belagiannis @bazilas.bsky.social · 29/03/2025

“Revisiting Gradient-Based Uncertainty for Monocular Depth Estimation” IEEE TPAMI, is here! We present a further formulation of our gradient-based method for quantifying #uncertainty in monocular #depth predictions #NoTrainingNeeded. Joint work between @fau.de & @uniulm.bsky.social. Links below.

100

Reposted by Vasileios Belagiannis

Andreas Geiger @andreasgeiger.bsky.social · 15/01/2025

Excited to share that today our paper recommender platform www.scholar-inbox.com has reached 20k users! We hope to reach 100k by the end of the year.. Lots of new features are being worked on currently and rolled out soon.

1219026

Reposted by Vasileios Belagiannis

Kosta Derpanis @csprofkgd.bsky.social · 05/01/2025

Interesting, it appears the #ICCV2025 submission and supplementary materials deadlines are the SAME.

3183

Vasileios Belagiannis @bazilas.bsky.social · 25/12/2024

Visual results

000

Vasileios Belagiannis @bazilas.bsky.social · 24/12/2024

Results: Extensive evaluation on ImageNet and CIFAR-10 datasets shows superior performance in filtering low-quality samples and improving generation quality. 📄 Paper: lnkd.in/d7JBSkiz 💻 Code: lnkd.in/dqFbC2nP Online demo coming soon! #FAU #MachineLearning #ComputerVision #DiffusionModels #WACV2025

lnkd.in

This link will take you to a page that’s not on LinkedIn

100

Vasileios Belagiannis @bazilas.bsky.social · 24/12/2024

💡 Theoretical Insights: We show that our uncertainty estimates are related to the second-order derivative of the diffusion noise distribution, providing a solid mathematical foundation.

100

Vasileios Belagiannis @bazilas.bsky.social · 24/12/2024

🎯 Guided sampling: We use uncertainty estimates to drive the sampling process towards higher quality generations, resulting in improved FID results and fewer artefacts.

100

Vasileios Belagiannis @bazilas.bsky.social · 24/12/2024

Key highlights: 📊 Training-free uncertainty: Our method estimates pixel-wise aleatoric uncertainty during the sampling phase without requiring any model modifications or additional training, allowing to filter out low quality samples.

100

Vasileios Belagiannis @bazilas.bsky.social · 24/12/2024

🎉 🎄 New WACV 2025 Publication! "Diffusion Model Guided Sampling with Pixel-Wise Aleatoric Uncertainty Estimation" by Michele De Vita, Vasileios Belagiannis @fau.de We're excited to introduce one of the first uncertainty estimation methods for diffusion models! Links & highlights below

110