CVPR 2026: DAAAM - Describe Anything, Anywhere, at Any Moment

Note: This repository is based on MIT-SPARK/DAAAM (BSD 3-Clause License — see LICENSE and THIRD_PARTY_NOTICES.md). It contains modifications for running the pipeline standalone (non-ROS) on the CODa dataset: rectified-intrinsics/extrinsics handling, a LiDAR (3d_comp) depth source with multi-sweep accumulation, and related scripts. See pipeline_and_results_summary.md.

[arXiv] [Project Page]

Real-time foundation-model-first robot mapping: SAM segmentation + BotSort tracking + VLM grounding feed into Hydra to build 3D Dynamic Scene Graphs on the fly.

Key contributions:

• Novel optimization-based frontend for semantic descriptions from localized captioning models
• Hierarchical 4D scene graph construction with real-time performance
• State-of-the-art results on NaVQA and SG3D benchmarks

The ROS 2 interface lives in DAAAM-ROS.

Installation | Running | Codebase | DAAAM-ROS

Paper

If you use this code in your work, please cite the following paper:

Nicolas Gorlo, Lukas Schmid, and Luca Carlone, "Describe Anything, Anywhere, at Any Moment". arXiv preprint arXiv:2512.00565, 2025.

@article{Gorlo2025DAAAM,
      title={Describe Anything Anywhere At Any Moment},
      author={Nicolas Gorlo and Lukas Schmid and Luca Carlone},
      year={2025},
      eprint={2512.00565},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2512.00565}
}

This work was supported by the ARL DCIST program and the ONR RAPID program.