NCore provides data representations, APIs, and tools to support data-driven neural reconstructions with a focus on robotics and autonomous vehicle data.
Project Site: research.nvidia.com/labs/sil/projects/ncore
- Neural Reconstruction - Designed for data-driven neural 3D reconstruction and simulation applications
- Data Format - Canonical and extensible component-based data format optimized for efficient training
- Sensor Models - GPU-accelerated Camera and LiDAR intrinsic models
- Converters - Transform external datasets into a common representation
pip install nvidia-ncoreFull documentation is available at nvidia.github.io/ncore.
Interested in contributing to NCore? See CONTRIBUTING.md for detailed instructions on setting up your development environment, coding guidelines, and how to submit your contributions.
This project will download and install additional third-party open source software as dependencies. Review the license terms of these open source projects before use. See ATTRIBUTIONS.md for a list of direct dependencies and their licenses.
NCore code-level bugs, documentation issues, and feature requests: file a GitHub issue using the appropriate template (Bug Report, Documentation Request, or Feature Request). The relevant NVIDIA responder is auto-assigned via the template's assignees: field.
Usage and how-to questions related to NuRec / Omniverse: please post on the NVIDIA Developer Forum (Omniverse / NuRec). Such questions are not tracked in this repository.
Security vulnerabilities: please use NVIDIA's Vulnerability Disclosure Program (see SECURITY.md). Do not file security issues publicly here.
NCore is provided under the Apache License, Version 2.0. See LICENSE for the full license text.