gradslam
An open-source differentiable dense SLAM library for PyTorch, enabling gradient flow from map outputs to sensor inputs.
What is gradslam?
gradslam is an open-source differentiable dense SLAM library for PyTorch that provides a framework for building SLAM systems where gradients can flow from outputs like maps and trajectories back to inputs such as raw RGB-D images and parameters. It solves the problem of integrating SLAM with deep learning by making traditional SLAM components differentiable, enabling optimization and end-to-end training. This allows researchers to develop and tune SLAM algorithms using gradient-based methods.
Researchers and developers in robotics, computer vision, and autonomous systems who are working on SLAM, 3D reconstruction, or neural scene representation and need differentiable geometric pipelines.
Developers choose gradslam because it uniquely provides a fully differentiable SLAM framework within PyTorch, enabling gradient-based optimization of SLAM systems and seamless integration with deep learning models for end-to-end training.
Overview
gradslam is an open source differentiable dense SLAM library for PyTorch
Use Cases
Best For
- Building differentiable SLAM systems for research and experimentation
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
