Coloring the Past: Neural Historical Buildings Reconstruction from Archival Photography David Komorowicz, Lu Sang, Ferdinand Maiwald, Daniel Cremers tl;dr: dense point clouds+images->SDF net; color appearance embedding loss->color synthetic view arxiv.org/pdf/2311.17810…

LEAP-VO: Long-term Effective Any Point Tracking for Visual Odometry Weirong Chen, Le Chen, Rui Wang, Marc Pollefeys tl;dr: gradient-max in grid as anchor->PIP-like tracker + modeling uncertainty -> Local BA -> good VO. arxiv.org/abs/2401.01887…

The COLMAP 3.9 release is out. Let's try it! github.com/colmap/colmap/…

can wild life be monitored & protected using autonomous #drone & #AI-based solutions? follow our WildDrone project & see the introductory video: youtube.com/watch?v=5m_3lk… #UAV #UAS Fondazione Bruno Kessler - FBK #Tech4Wildlife #Conservationtechnology #vision #photogrammetry #mapping

Folks, who work in image matching, 3D reconstruction, NERFs, etc and would like to help us with reviewing papers for IMW2024 - please DM me. The load is 1-3 papers per person. #CVPR2024

Submit your image matching and 3D-related papers to Image Matching Workshop 2024. Submission deadline: March 18 Decision: April 8. Paper size: 8 pages cmt3.research.microsoft.com/IMW2024 #CVPR2026 #CVPR2024

Our workshop "Equivariant Vision: From Theory to Practice" will be hosted at #CVPR2024 in Seattle this summer! #CVPR2026 Both original and published works are welcome to submit to our workshop! 🔗equivision.github.io ⏰Deadline: Mar 22, 2024

Image #matching and extraction of correspondences are the core of the #photogrammetry pipeline. We use template matching to improve #keypoint locations and achieve sub-pixel #accuracy results link.springer.com/chapter/10.100… #heritage #vision @ICIAP2023

Fully-funded PhD position open: My friend and colleague from FEL, CTU in Prague is looking for a PhD student for topic "Learning Quantized Neural Networks and Discrete Representations" euraxess.ec.europa.eu/jobs/186695 Deadline: Feb 29 Supervisor and questions to: cmp.felk.cvut.cz/~shekhovt/

📣 Dev news! ⬇️ 📷#DeepImageMatching, our multi-view matcher with hand-crafted and #DL features, hit its first stable release v0.1.0! Now with documentation, notebook support, and Colab examples. Next up: exporting matches to #OpenMVG and #MicMac. Stay tuned! Luca Morelli 3DOMFBK

We are looking for new collaborators for expanding DIM 🔼 We need to improve code efficiency, implement new matching approaches and improve the dense matching with #LOFTR and #RoMa. And we always need testers! Feel free to contact us (or open a PR on Github)

📣 Nice presentation of #DeepImageMatching by Luca Morelli (3DOMFBK) at ISPRS 3D-Arch Siena (Italy). 📷 Luca presented our tool for multi-view image matching that allows for using both hand-crafted and DL features with a single easy-to-use interface. 🔗github.com/3DOM-FBK/deep-…

10th edition of the ISPRS CIPA Heritage Documentation 3D-ARCH workshop 3darch.fbk.eu with more than 120 registrations & some 65 papers isprs-archives.copernicus.org/articles/XLVII… #heritage #photogrammetry #3D #VR #AR Fondazione Bruno Kessler - FBK Politecnico di Milano Università di Siena

Dmytro Mishkin 🇺🇦 MrNeRF If someone is interested in getting close to linear complexity for incremental SfM, this is a good read: ccwu.me/vsfm/vsfm.pdf

a tool to find image corresponding using different strategies, from hand-crafted to #DeepLearning methods. Notebook support. #Colab examples. Export for #Colmap, #OpenMVG & #MicMac isprs-archives.copernicus.org/articles/XLVII… github.com/3DOM-FBK/deep-… Francesco Ioli Luca Morelli #photogrammetry

We are launching Structured Semantic 3D Reconstruction (S23DR) Challenge at #CVPR2026 usm3d workshop on Hugging Face! Prizes: 1: $10,000 2: $7,000 3: $5,000 Deadline: June 4, 2024 Task: point cloud+segmentations+monodepth ->roof wireframe. huggingface.co/spaces/usm3d/S… 1/2

new release of the Deep-Image-Matching framework: github.com/3DOM-FBK/deep-… - export for #OpenMVG & #MicMac - different camera models - Graph visualization (thanks @gperda) - Docker installation (thanks Kauê) Francesco Ioli Luca Morelli #photogrammetry #AI #deeplearning

Exploring the potential of DL local features for co-registering multi-temporal image blocks from different sensors. 📸 Here's a comparison of thermal vs. RGB images. Big thanks to @nneilsutherland for sharing this interesting dataset! 👏

Given the 3DV conference being super relevant to everyone in image matching, we moved Image Matching Workshop 2024 paper submission deadline to March 25 2024. So you still have time until Monday to submit your papers! cmt3.research.microsoft.com/IMW2024/Submis… #CVPR2026

** 50 years of #SfM ** join our European Conference on Computer Vision #ECCV2026 workshop in #Milan 50sfm.fbk.eu 7 excellent invited speakers & challenging sequences #photogrammetry #3D #imagematching #AI #SLAM @fusiello Rongjun Qin Fondazione Bruno Kessler - FBK Marc Pollefeys pix4d Luca Carlone Dmytro Mishkin 🇺🇦 Noah Snavely Luca Morelli