Antoni Rosinol

Email:

Citations: 792 • h-index: 6 • i10-index: 6
Resume • GitHub • Google Scholar • Linkedin • Twitter

News

[Oct 3, 2022] New WACV-accepted paper: Probabilistic Volumetric Fusion for Dense Monocular SLAM. Dense and accurate 3D mesh reconstruction of scenes in real-time from video!
[May 4, 2022] Invited Talk at UPC, Barcelona.
[Apr 26, 2022] I was awarded the prestigious Rafael del Pino Scholarship!
[Apr 6, 2022] It was a great pleasure to present Kimera + 3D DSGs as an invited guest lecture for Prof. Ghaffari’s excellent course on MOBILE ROBOTICS. I also shared my thoughts about the Future of SLAM: i.e. tightly-coupling learning and SLAM + Differentiable Rendering. Btw, Maani’s lectures are a gem :). Video
[Oct 17, 2021] Our workshop on Structural and Compositional Learning on 3D Data at ICCV was a great success! Thank you for attending! Here are all videos: Part A, Part B.
[Sept 30, 2021] I will be co-chair of the Mapping II session at IROS this year, looking forward to it!
[Apr 2, 2021] Gave an invited talk at Magic Leap on “Kimera: from SLAM to Spatial Perception with 3D Dynamic Scene Graphs”.
[Mar 9, 2021] Gave an invited talk at the Robotics and Perception Group @ UZH on our Kimera journal paper!
[Jan 1, 2021] At last! I just sent our IJRR Special Issue Journal paper: Wrapping up everything together and running in real-life scenarios is very very difficult, but, finally, here I present the work unifying all that Kimera represents: a Spatial Perception engine that opens new and exciting research directions.
[Oct 6, 2021] We were invited to present an extended version of our 3D DSG paper at IJRR for the RSS2020 Special Issue!

[expand]

[July 14, 2020] Presented 3D Dynamic Scene Graphs (DSG) @ RSS. In this video I show the structure of a 3D DSG, how we build it from visual-inertial data, and how we can use it for semantic path planning (with an actual query at the end of the video!).
[Jun 12, 2020] Presented Kimera @ ICRAxMIT.
[May 1, 2020] Checkout our Hands-On Tutorial on Kimera!
[Feb 18, 2020] Excited to present our newest work: 3D Dynamic Scene Graphs, a unified representation for actionable spatial perception.
[Dec 5, 2019] The metric-semantic reconstruction in Kimera is now much faster! Checkout the results here.
[Sep 20, 2019] Open-Source release of Kimera! An accurate visual-inertial odometry pipeline with SLAM capabilities and multi-horizon 3D Mesh generation with Semantic annotations. Check https://github.com/MIT-SPARK/Kimera.
[June 10, 2019] Presented our new work Kimera, a metric-semantic SLAM pipeline, at the Northeast Robotics Colloquium (NERC) 2019 at the University of Pennsylvania!
[May 24, 2019] IEEE RAL’18 Best Paper Award Honourable Mention: Our paper UltimateSLAM received the IEEE Robotics and Automation Letters 2018 Best Paper Award Honourable Mention during the award session at the ICRA 2019 conference in Montreal. It ranked in the top 3 out of 520 papers published by RAL in 2018. The project’s website can be found here. [Paper][Video].
[May 23, 2019] Presented our recent work Towards Real-Time Metric-Semantic Localization and Mapping with 3D Mesh Models at the 2nd International Workshop on Lines, Planes and Manhattan Models for 3-D Mapping (LPM 2019). We also show our newest 3D photo-realistic simulator and present our vision for the future of SLAM. [Slides].
[Jan 26, 2019] Our paper Incremental Visual-Inertial 3D Mesh Generation with Structural Regularities accepted at ICRA 2019. We show how to use structural regularities in the scene to improve localization and mapping performance in real-time. The project’s website can be found here. [Paper][Video]

[/expand]

Publications

Journal Papers

Kimera: from SLAM to Spatial Perception with 3D Dynamic Scene Graphs

A. Rosinol, A. Violette, M. Abate, N. Hughes, Y. Chang, J. Shi, A. Gupta, L. Carlone

Intl. J. of Robotics Research (2021)

bibtex

@article{Rosinol21ijrr-Kimera,
  title = {Kimera: from {SLAM} to Spatial Perception with {3D} Dynamic Scene Graphs},
  author = {Rosinol, A. and Violette, A. and Abate, M. and Hughes, N. and Chang, Y. and Shi, J. and Gupta, A. and Carlone, L.},
  journal = {Intl. J. of Robotics Research},
  volume = {40},
  number = {12--14},
  pages = {1510--1546},
  note = {arXiv preprint arXiv: 2101.06894, \linkToPdf{https://arxiv.org/pdf/2101.06894.pdf}},
  pdf = {https://arxiv.org/pdf/2101.06894.pdf},
  year = {2021}
}

Ultimate SLAM? Combining events, images, and IMU for robust visual SLAM in HDR and high-speed scenarios

A. Rosinol, H. Rebecq, T. Horstschaefer, D. Scaramuzza

IEEE Robotics and Automation Letters (2018)

bibtex pdf doi link video

@article{Rosinol18ral-ultimate,
  title = {{U}ltimate {SLAM}? {C}ombining events, images, and {IMU} for robust visual {SLAM} in {HDR} and high-speed scenarios},
  author = {Rosinol, A. and Rebecq, H. and Horstschaefer, T. and Scaramuzza, D.},
  journal = {IEEE Robotics and Automation Letters},
  volume = {3},
  number = {2},
  doi = {10.1109/LRA.2018.2793357},
  video = {https://www.youtube.com/watch?v=jIvJuWdmemE},
  pdf = {https://arxiv.org/abs/1709.06310},
  pages = {994--1001},
  year = {2018},
  publisher = {IEEE},
  url = {http://rpg.ifi.uzh.ch/ultimateslam.html},
  award = {Best Paper Award 
 Finalist},
  note = {\linkToPdf{https://arxiv.org/abs/1709.06310},
            \linkToWeb{http://rpg.ifi.uzh.ch/ultimateslam.html}}
}

Best Paper Award
Finalist

Conference Papers

Probabilistic Volumetric Fusion for Dense Monocular SLAM

A. Rosinol, J. Leonard, L. Carlone

IEEE Workshop on Applications of Computer Vision (WACV) (2022)

bibtex pdf
```
@inproceedings{Rosinol22wacv,
  title = {{Probabilistic Volumetric Fusion for Dense Monocular SLAM}},
  author = {Rosinol, A. and Leonard, J. and Carlone, L.},
  booktitle = {IEEE Workshop on Applications of Computer Vision (WACV)},
  pdf = {https://arxiv.org/pdf/2210.01276.pdf},
  note = {\linkToPdf{https://arxiv.org/pdf/2210.01276.pdf}},
  year = {2022}
}
```
Smooth Mesh Estimation from Depth Data using Non-Smooth Convex Optimization

A. Rosinol, L. Carlone

IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) (2021)

bibtex
```
@inproceedings{Rosinol21iros-mesh,
  title = {Smooth Mesh Estimation from Depth Data using Non-Smooth Convex Optimization},
  author = {Rosinol, A. and Carlone, L.},
  booktitle = {IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS)},
  year = {2021},
  pdf = {https://arxiv.org/pdf/2108.02957.pdf},
  note = {\linkToPdf{https://arxiv.org/pdf/2108.02957.pdf}}
}
```

3D Dynamic Scene Graphs: Actionable Spatial Perception with Places, Objects, and Humans

A. Rosinol, A. Gupta, M. Abate, J. Shi, L. Carlone

Robotics: Science and Systems (RSS) (2020)

bibtex pdf doi link video

@inproceedings{Rosinol20rss-dsg,
  title = {{3D} Dynamic Scene Graphs: Actionable Spatial Perception with Places, Objects, and Humans},
  author = {Rosinol, A. and Gupta, A. and Abate, M. and Shi, J. and Carlone, L.},
  booktitle = {Robotics: Science and Systems (RSS)},
  year = {2020},
  pdf = {https://arxiv.org/pdf/2002.06289.pdf},
  url = {http://news.mit.edu/2020/robots-spatial-perception-0715},
  video = {https://www.youtube.com/watch?v=SWbofjhyPzI},
  doi = {10.15607/RSS.2020.XVI.079},
  note = {\linkToPdf{https://arxiv.org/pdf/2002.06289.pdf},
            \linkToMedia{http://news.mit.edu/2020/robots-spatial-perception-0715},
            \linkToVideo{https://www.youtube.com/watch?v=SWbofjhyPzI&feature=youtu.be}}
}

Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping

A. Rosinol, M. Abate, Y. Chang, L. Carlone

IEEE Intl. Conf. on Robotics and Automation (ICRA) (2020)

bibtex pdf doi code video

@inproceedings{Rosinol20icra-Kimera,
  title = {Kimera: an Open-Source Library for Real-Time Metric-Semantic Localization and Mapping},
  author = {Rosinol, A. and Abate, M. and Chang, Y. and Carlone, L.},
  booktitle = {IEEE Intl. Conf. on Robotics and Automation (ICRA)},
  year = {2020},
  pdf = {https://arxiv.org/pdf/1910.02490.pdf},
  video = {https://www.youtube.com/watch?v=-5XxXRABXJs},
  code = {https://github.com/MIT-SPARK/Kimera},
  doi = {10.1109/ICRA40945.2020.9196885},
  note = {arXiv preprint arXiv: 1910.02490,
            \linkToVideo{https://www.youtube.com/watch?v=-5XxXRABXJs},
            \linkToCode{https://github.com/MIT-SPARK/Kimera},
            \linkToPdf{https://arxiv.org/pdf/1910.02490.pdf}}
}

Incremental Visual-Inertial 3D Mesh Generation with Structural Regularities

A. Rosinol, T. Sattler, M. Pollefeys, L. Carlone

IEEE Intl. Conf. on Robotics and Automation (ICRA) (2019)

bibtex pdf doi link video

@inproceedings{Rosinol19icra-mesh,
  author = {Rosinol, A. and Sattler, T. and Pollefeys, M. and Carlone, L.},
  title = {Incremental {V}isual-{I}nertial {3D} {M}esh {G}eneration with {S}tructural {R}egularities},
  booktitle = {IEEE Intl. Conf. on Robotics and Automation (ICRA)},
  year = {2019},
  pdf = {https://arxiv.org/pdf/1903.01067.pdf},
  video = {https://www.youtube.com/watch?v=C5fFDEJ9cFQ},
  url = {https://www.mit.edu/\%7Earosinol/research/struct3dmesh.html},
  doi = {10.1109/ICRA.2019.8794456},
  note = {\linkToPdf{https://arxiv.org/pdf/1903.01067.pdf},
            \linkToWeb{https://www.mit.edu/~arosinol/research/struct3dmesh.html}}
}

Densifying Sparse VIO: a mesh-based approach using Structural Regularities

A. Rosinol

(2018)

bibtex pdf doi

@mastersthesis{Rosinol18thesis,
  author = {Rosinol, A.},
  title = {{D}ensifying {S}parse {VIO}: a mesh-based approach using {S}tructural {R}egularities},
  year = {2018},
  language = {en},
  copyright = {Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International},
  keywords = {Visual Inertial Odometry; Dense Mapping; State Estimation; SLAM; Robotics},
  doi = {10.3929/ethz-b-000297645},
  school = {ETH Zurich},
  pdf = {https://www.research-collection.ethz.ch/handle/20.500.11850/297645},
  note = {\linkToPdf{https://www.research-collection.ethz.ch/handle/20.500.11850/297645}}
}

Antoni Rosinol

News

Publications

Journal Papers

Conference Papers

Code

Datasets