Point Cloud Library

Point Cloud Library (PCL) is a free, open-source framework for processing 2D and 3D images and point clouds. Its algorithms cover filtering, feature estimation, surface reconstruction, registration, model fitting, and segmentation. These tools can reduce noise, join point clouds, divide scenes into segments, locate keypoints, calculate descriptors, create surfaces, and visualize point data. PCL is divided into smaller code libraries that can be compiled separately, supporting distribution on systems with limited computing capacity or storage. Prebuilt binaries are available for Linux, Windows, and macOS; the project recommends package managers to simplify distribution and installation of third-party dependencies. PCL supports OpenNI 3D interfaces for acquiring and processing data from devices such as PrimeSensor cameras, Microsoft Kinect, and Asus XTionPro. Its API reference includes public methods and classes, code examples, and links to scientific publications about methods. PCL uses the 3-clause BSD license and is free for commercial and research use. The project directs general usage questions to Stack Overflow and its Discord server.

PCL provides a broad set of point-cloud and image-processing algorithms under a free BSD license. Installation involves third-party dependencies, and the project points users to community channels for general questions.

PCL suits developers and researchers working with 2D or 3D images and point-cloud data. Its separate libraries may also suit projects with limited computing capacity or storage.

Point Cloud Library (PCL) is a modular framework for developers building 2D and 3D image and point-cloud processing into software. It is best for teams that need a wide algorithm toolkit and can handle a development-oriented setup; it is not a ready-made point-cloud application.

Overview

PCL combines filtering, feature estimation, surface reconstruction, registration, model fitting, and segmentation. Together, these support workflows such as cleaning noisy scans, stitching point clouds, dividing scenes into segments, extracting keypoints, computing descriptors, generating surfaces, and visualizing results. That range makes PCL a capable foundation for varied processing pipelines, but users who want a finished interface rather than a library should choose another tool.

The project is open source under the 3-clause BSD license and is free for commercial and research use. Its modular design lets developers compile smaller libraries separately, which can help when distributing software to systems with tighter compute or storage constraints. PCL also supports OpenNI 3D interfaces for acquiring and processing data from devices such as PrimeSensor cameras, Microsoft Kinect, and Asus XTionPro.

Key features

• Broad algorithm coverage: Filtering, feature estimation, surface reconstruction, registration, model fitting, and segmentation cover many core point-cloud operations in one framework. This breadth suits developers assembling custom workflows; it still requires them to build the application around the algorithms.
• Modular compilation: Separately compilable libraries can help reduce what a distributed application carries on systems with limited compute or storage capacity.
• Sensor interfaces: Native OpenNI support serves teams acquiring and processing data from compatible devices, including PrimeSensor cameras, Microsoft Kinect, and Asus XTionPro.
• Developer documentation: The API reference explains public methods and classes, includes code snippets, and links to scientific publications about the methods. General usage questions are directed to Stack Overflow and the project's Discord community.

Pricing

PCL costs nothing, with no licensing fee for commercial or research use under its BSD license. There is no free trial because the software is free. The trade-off is setup work: PCL depends on third-party libraries, and the project recommends package managers to simplify installation. Prebuilt binaries are available for Linux, Windows, and macOS.

Platforms

PCL supports Android, Linux, macOS, and Windows. Prebuilt binaries are available for Linux, Windows, and macOS; the project recommends package managers as its preferred distribution method. Teams targeting Android should account for the modular design, which allows smaller libraries to be compiled separately, and assess the dependency setup for their build.

Who it's for

PCL is a strong fit for developers and research teams building software that needs point-cloud registration, classification, processing, or point-cloud-to-mesh workflows. Its range of algorithms and commercial-use-friendly license make it practical for projects that need to adapt processing components to their own pipelines. It is a weaker fit for people who want a turnkey desktop product, or for teams that cannot accommodate third-party dependencies and community-based support.

Pros and cons

Pros

• Wide processing toolkit: Filtering, registration, segmentation, feature estimation, model fitting, and surface reconstruction support varied workflows without requiring a separate framework for each operation.
• Permissive commercial use: The 3-clause BSD license allows free use in commercial as well as research projects.
• Build flexibility: Compiling smaller libraries separately can help teams target devices with constrained compute or storage.
• Method-level documentation: API explanations, snippets, and publication links are useful to developers who need to understand and implement specific algorithms.

Cons

• Dependency overhead: Third-party libraries are required, so installation is more involved than choosing a self-contained end-user application.
• Community-led general support: Users are pointed to Stack Overflow and Discord for usage questions rather than a dedicated support channel.
• Security caution: The repository README warns that a copy of the old website was hacked and may still host malicious code; users should avoid that old site copy.

Alternatives

Point Cloud Software is a useful directory for comparing other tools in this category.

• Global Mapper is a paid option with a free trial; its Essentials plan starts at $599 a year and includes data viewing and format conversion plus coordinate transformation and reprojection. Consider it when those stated capabilities matter more than a free, developer-oriented library.
• WhiteboxTools offers a free open-core package with 700+ tools for Python, R, or QGIS, as well as paid LiDAR and forest analytics. Choose it if that tool count or those integrations better match your workflow.
• LidarView is a free, open-source alternative under Apache 2.0 for Linux and Windows. It may suit users looking for a free option on those platforms.
• LP360 has a free plan and paid options, including DJI L1/L2 cloud batch processing for $620 a year, with a 20-flight monthly limit and one user included. Consider it when that specific batch-processing plan fits your needs.
• Leica Cyclone 3DR is another option for Windows users.
• Open3D is a free alternative available across Linux, macOS, Windows, web, API, and self-hosted platforms. Consider it when those deployment options are a better match.
• Pointly uses freemium pricing and offers usage-based processing priced from 0.03 € per million points, with volume booked in advance and unlimited parallel jobs. It may suit users who prefer usage-based pricing for that workflow.
• Entwine is another free option for Linux users.

Verdict

Choose PCL if you are building a custom point-cloud or image-processing workflow and want a broad algorithm foundation that can be used commercially without licensing fees. Its strongest case is the combination of processing breadth, modular compilation, and BSD licensing. Look elsewhere if you need a finished application, want to avoid third-party setup, or require support beyond community channels.