Over the past two decades, research in the field of Simultaneous Localization and Mapping (SLAM) has undergone a significant evolution, highlighting its critical role in enabling autonomous exploration of unseen environments. This evolution ranges from hand-crafted methods, through the era of deep learning, to more recent developments in the wake of Neural Radiance Fields (NeRFs) and 3D Gaussian Splatting (3DGS) representations.
These latter advancements, in particular, are happening right now at unprecedented speed: indeed, since the first NeRF-inspired SLAM system came out in 2021 -- iMAP: Implicit Mapping and Positioning in Real-Time, ICCV 2021 -- more than 200 papers have been posted on arXiv or published at top-tier conferences in Computer Vision and Machine Learning. For a recent overview, see Tosi et al. "How NeRFs and 3D Gaussian Splatting are Reshaping SLAM: a Survey".
This sudden revolution brought excitement to the research community paving the way, on the one hand, for new applications beyond the original purpose of SLAM systems. Among these is the possibility of rendering realistic novel views from unobserved viewpoints, something made possible by the recent advances in novel view synthesis, which further extends the overall SLAM purpose.
On the other hand, new challenges emerged caused by the speed at which newer and newer SLAM systems are being proposed.
For example, as many papers came out in a very short amount of time, most of them are concurrent and did not have the chance to compare to each other, leaving several open questions -- e.g., which are the real key factors in many of these SLAM systems? Which scene representation is the best suited?
The workshop aims to bring researchers together being interested in dense neural SLAM methods and provide a forum for the mutual exchange of experiences, beliefs, intuitions, ideas and future research directions.
With a mix of invited keynote talks of renown experts, a poster session and a panel discussion, we invite people both from academia and industry across all seniority levels to learn from each other and engage in discussions.
Federico Tombari
Director of Research
Google Zurich &
Lecturer at TUM
Jakob Engel
Director of Research
Meta Reality Labs
Luca Carlone
Associate Professor
MIT
Angjoo Kanazawa
Assistant Professor
UC Berkeley
13:00 Introduction by organizers
13:05 Keynote talk #1: Luca Carlone, MIT
13:40 Keynote talk #2: Angjoo Kanazawa, UC Berkeley
14:15 Poster Session & Coffee break
15:15 Keynote talk #3: Federico Tombari, Google & TUM
15:50 Keynote talk #4: Jakob Engel, Meta
16:25 Panel discussion
17:00 Closing remarks
DROID-Splat Combining end-to-end SLAM with 3D Gaussian Splatting
Christian Homeyer, León Begiristain, Christoph Schnoerr, NeuSLAM ICCV Workshop 2025
MUSt3R: Multi-view Network for Stereo 3D Reconstruction
Yohann Cabon, Lucas Stoffl, Leonid Antsfeld, Gabriela Csurka, Boris Chidlovskii, Jerome Revaud, Vincent Leroy, CVPR 2025
Feed-Forward SceneDINO for Unsupervised Semantic Scene Completion
Aleksandar Jevtić, Christoph Reich, Felix Wimbauer, Oliver Hahn, Christian Rupprecht, Stefan Roth, Daniel Cremers, ICCV 2025
DeGauss: Dynamic-Static Decomposition with Gaussian Splatting for Distractor-free 3D Reconstruction
Rui Wang, Quentin Lohmeyer, Mirko Meboldt, Siyu Tang, ICCV 2025
PseudoMapTrainer: Learning Online Mapping without HD Maps
Christian Löwens, Thorben Funke, Jingchao Xie, Alexandru Paul Condurache, ICCV 2025
Splat-LOAM: Gaussian Splatting LiDAR Odometry and Mapping
Emanuele Giacomini, Luca Di Giammarino, Lorenzo De Rebotti, Giorgio Grisetti, Martin R. Oswald, ICCV 2025
Back on Track: Bundle Adjustment for Dynamic Scene Reconstruction
Weirong Chen, Ganlin Zhang, Felix Wimbauer, Rui Wang, Nikita Araslanov, Andrea Vedaldi, Daniel Cremers, ICCV 2025
ToF-Splatting: Dense SLAM using Sparse Time-of-Flight Depth and Multi-Frame Integration
Andrea Conti, Matteo Poggi, Valerio Cambareri, Martin R. Oswald, Stefano Mattoccia, ICCV 2025
Humans as a Calibration Pattern: Dynamic 3D Scene Reconstruction from Unsynchronized and Uncalibrated Videos
Changwoon Choi, Jeongjun Kim, Geonho Cha, Minkwan Kim, Dongyoon Wee, Young Min Kim, ICCV 2025
SceneSplat: Gaussian Splatting-based Scene Understanding with Vision-Language Pretraining
Yue Li, Qi Ma, Runyi Yang, Huapeng Li, Mengjiao Ma, Bin Ren, Nikola Popovic, Nicu Sebe, Ender Konukoglu, Theo Gevers, Luc Van Gool, Martin R. Oswald, Danda Pani Paudel, ICCV 2025
Vivid4D: Improving 4D Reconstruction from Monocular Video by Video Inpainting
Jiaxin Huang, Sheng Miao, BangBang Yang, Yuewen Ma, Yiyi Liao, ICCV 2025