Afterwards, please stop by to see MARCO, our approach for semantic correspondence that generalizes to novel keypoints and categories. 4:12 PM in oral session 4D and later at poster #20. #CVPR2026 github.com/visinf/MARCO

Come see INSID3, in-context segmentation with DINOv3, at 2:00 PM in oral session 4D and later at poster #19. #CVPR2026

Gabriele Trivigno@gabtriv.bsky.social · 4mo ago

🔥 Can in-context segmentation emerge directly from frozen DINOv3 features? At #CVPR2026, we present INSID3: Training-Free In-Context Segmentation with DINOv3 — a collaboration between PoliTo, TU Darmstadt and TU Munich. Check it out: github.com/visinf/INSID3

Our #CVPR2026 poster starts in 30 minutes! Come by Poster Session 2, #333 to learn about VideoCUPS — the first unsupervised video panoptic segmentation method, learning to detect, segment, and track objects in video without human supervision.

Visual Inference Lab@visinf.bsky.social · 2mo ago

📢 [CVPR’26] Can we learn to detect, segment, and track every object in a video without human supervision?  Yes, we introduce VideoCUPS, the first unsupervised video panoptic segmentation (VPS) method: 1. Get pseudo-labels from monocular videos. 2. Train a VPS model on them.

[1/6] 📢 We are in Denver at #CVPR2026 presenting 5 papers!

I am one of the potential advisors for this PhD position in our #DAAD funded AI graduate school ELIZA (eliza.school), which connects seven #ELLIS units in Germany. If you are interested in working with one of my colleagues or with me (in computer vision & deep learning), please consider applying.

Zuse School ELIZA@zuseschooleliza.bsky.social · 8mo ago

Are you looking for a PhD in machine learning? 🧠 ELIZA is hiring a Doctoral Researcher at TU Darmstadt! Apply by Jan 5: www.career.tu-darmstadt.de/tu-darmstadt... ℹ️ Learn more about ELIZA: eliza.school @tuda.bsky.social, @cs-tudarmstadt.bsky.social, @ellis.eu, @daadworldwide.bsky.social #AI

📢🎓 We have open PhD positions in Computer Vision & Machine Learning at @tuda.bsky.social and @hessianai.bsky.social within the Reasonable AI Cluster of Excellence — supervised by @stefanroth.bsky.social, @simoneschaub.bsky.social and many others! www.career.tu-darmstadt.de/tu-darmstadt...

📢 #CVPR2025 Highlight: Scene-Centric Unsupervised Panoptic Segmentation 🔥 We present CUPS, the first unsupervised panoptic segmentation method trained directly on scene-centric imagery. Using self-supervised features, depth & motion, we achieve SotA results! 🌎 visinf.github.io/cups

Starting in about 30 minutes in East A-C

Visual Inference Lab@visinf.bsky.social · 2y ago

Want to learn about how model design choices affect the attribution quality of vision models? Visit our #NeurIPS2024 poster on Friday afternoon (East Exhibition Hall A-C #2910)! Paper: arxiv.org/abs/2407.11910 Code: github.com/visinf/idsds