Inclusion Request for Customizable SLAM Robustness Benchmark (NeRF vs Gaussian Splatting representation for instant 3D reconstruction and localization)

[Xiaohao Xu]

Hey guys,

We sincerely appreciate the value of your repository! We would like to kindly request the inclusion of our customizable SLAM robustness benchmark, which incorporates a noisy data simulation pipeline for a comprehensive evaluation of SLAM performance. Our benchmark mainly focuses on evaluating the robustness of the latest monocular/multi-modal Neural-based SLAM models, including NeRF-based and Gaussian-Splatting-based SLAM, which reveals some intriguing properties of NeRF and GS representation for instant 3D reconstruction and localization. Additionally, it offers a preliminary assessment of multi-view and multi-agent/collaborative SLAM robustness.

We are confident that the addition of our benchmark will contribute to making SLAM more deployable and robust. We greatly appreciate your consideration. Thanks!

• Customizable-SLAM: Customizable Perturbation Synthesis for Robust SLAM Benchmarking, ArXiV, 2024. [Paper] [Code] [Video]

[Andy Baker]

Instructions for submitting pull requests are here: https://github.com/awesome-NeRF/awesome-NeRF/blob/main/how-to-PR.md

[Xiaohao Xu]

Got that. Thanks!