Open3D is an open-source library for 3D data processing, designed to support rapid development of software that deals with 3D data. It provides a set of carefully selected data structures and algorithms in both C++ and Python, with a highly optimized and parallelized backend. Open3D was developed from scratch with a small and carefully considered set of dependencies, making it easy to set up on different platforms and compile from source with minimal effort. The code is clean, consistently styled, and maintained via a clear code review mechanism. Open3D has been used in several published research projects and is actively deployed in the cloud. The library is released under the MIT license and is available at http://www.open3d.org. Open3D was created to address the need for a fast, easy-to-use, and modern open-source library for 3D data processing. It supports three main data representations: point clouds, meshes, and RGB-D images. It provides a complete set of basic processing algorithms for each representation, including I/O, sampling, visualization, and data conversion. It also includes a collection of widely used algorithms, such as normal estimation, ICP registration, and volumetric integration. Open3D has been used to implement complete workflows such as large-scale scene reconstruction. Open3D provides a Python binding that allows developers to use Python as a glue language to assemble components implemented in the backend. It has a simple and intuitive API, and its Python implementation is significantly shorter and clearer than the C++ implementation. Open3D has nine modules, including data, visualization, registration, reconstruction, and utility functions. It supports a wide range of 3D data formats and provides functions for reading and writing 3D data files. Open3D is optimized for high performance, with a highly optimized C++ backend that is set up for parallelization. It has been shown to be faster than other 3D processing libraries, with some implementations up to 25 times faster than their counterparts in PCL. Many functions are parallelized using OpenMP, which allows for efficient processing of large datasets. Open3D is released under the MIT license and is available at http://www.open3d.org. It is actively developed and maintained by a community of open-source contributors.Open3D is an open-source library for 3D data processing, designed to support rapid development of software that deals with 3D data. It provides a set of carefully selected data structures and algorithms in both C++ and Python, with a highly optimized and parallelized backend. Open3D was developed from scratch with a small and carefully considered set of dependencies, making it easy to set up on different platforms and compile from source with minimal effort. The code is clean, consistently styled, and maintained via a clear code review mechanism. Open3D has been used in several published research projects and is actively deployed in the cloud. The library is released under the MIT license and is available at http://www.open3d.org. Open3D was created to address the need for a fast, easy-to-use, and modern open-source library for 3D data processing. It supports three main data representations: point clouds, meshes, and RGB-D images. It provides a complete set of basic processing algorithms for each representation, including I/O, sampling, visualization, and data conversion. It also includes a collection of widely used algorithms, such as normal estimation, ICP registration, and volumetric integration. Open3D has been used to implement complete workflows such as large-scale scene reconstruction. Open3D provides a Python binding that allows developers to use Python as a glue language to assemble components implemented in the backend. It has a simple and intuitive API, and its Python implementation is significantly shorter and clearer than the C++ implementation. Open3D has nine modules, including data, visualization, registration, reconstruction, and utility functions. It supports a wide range of 3D data formats and provides functions for reading and writing 3D data files. Open3D is optimized for high performance, with a highly optimized C++ backend that is set up for parallelization. It has been shown to be faster than other 3D processing libraries, with some implementations up to 25 times faster than their counterparts in PCL. Many functions are parallelized using OpenMP, which allows for efficient processing of large datasets. Open3D is released under the MIT license and is available at http://www.open3d.org. It is actively developed and maintained by a community of open-source contributors.