PCL 1.15.0 brings:

• a torus model for RANSAC
• several classes are now parallelized, including PrincipalCurvaturesEstimation, RadiusOutlierRemoval, ICP, and parts of GICP
• compatibility with the newest compilers, Boost versions, etc
• and of course many bug fixes, speed improvements, ...

For an exhaustive list of newly added features, deprecations and other changes in PCL 1.15.0, please see CHANGES.md.
Thanks to everyone who contributed to this release!

PCL 1.15.0 brings:

• a torus model for RANSAC
• several classes are now parallelized, including PrincipalCurvaturesEstimation, RadiusOutlierRemoval, ICP, and parts of GICP
• compatibility with the newest compilers, Boost versions, etc
• and of course many bug fixes, speed improvements, ...

For an exhaustive list of newly added features, deprecations and other changes
in PCL 1.15.0, please see CHANGES.md.

One big change in this release is that the dependency boost-filesystem is now optional if you compile PCL and your own project that uses PCL as C++17 (e.g. by adding -DCMAKE_CXX_STANDARD=17 -DCMAKE_CUDA_STANDARD=17 to the cmake call). The default C++ standard for PCL is currently still C++14, where boost-filesystem is a hard requirement. Some modules will not be built if boost-filesystem is not available (currently recognition, outofcore, some apps). Big thanks to @cybaol for implementing this, and of course also thanks to everyone else who contributed to this release!

For an exhaustive list of newly added features, deprecations and other changes
in PCL 1.14.1, please see CHANGES.md.

One big change in this release is that the dependency boost-filesystem is now optional if you compile PCL and your own project that uses PCL as C++17 (e.g. by adding -DCMAKE_CXX_STANDARD=17 -DCMAKE_CUDA_STANDARD=17 to the cmake call). The default C++ standard for PCL is currently still C++14, where boost-filesystem is a hard requirement. Some modules will not be built if boost-filesystem is not available (currently recognition, outofcore, some apps). Big thanks to @cybaol for implementing this, and of course also thanks to everyone else who contributed to this release!

For an exhaustive list of newly added features, deprecations and other changes
in PCL 1.14.1, please see CHANGES.md.

This release brings for example:

• a faster and more robust GeneralizedIterativeClosestPoint (GICP) registration
• better compatibility with the latest compilers and 3rd party libraries (Eigen, Boost, etc.)
• and many more bug fixes and performance enhancements

For an exhaustive list of newly added features, deprecations and other changes in PCL 1.14.0, please see CHANGES.md.

This release brings for example:

• a faster and more robust GeneralizedIterativeClosestPoint (GICP) registration
• better compatibility with the latest compilers and 3rd party libraries (Eigen, Boost, etc.)
• and many more bug fixes and performance enhancements

For an exhaustive list of newly added features, deprecations and other changes in PCL 1.14.0, please see CHANGES.md.

What's Changed

This release has several new features and bug fixess. To name a few:

• [common] Add overload to fromPCLPointCloud2 to avoid copying data [#5608]
• [io] Add writeBinary to ostream for PCDWriter [#5598]

For an exhaustive list of newly added features, deprecations and other changes in PCL 1.13.1, please see CHANGES.md.

Thanks to everyone who contributed to this release!

What's Changed

This release has several new features and bug fixess. To name a few:

• [common] Add overload to fromPCLPointCloud2 to avoid copying data [#5608]
• [io] Add writeBinary to ostream for PCDWriter [#5598]

For an exhaustive list of newly added features, deprecations and other changes in PCL 1.13.1, please see CHANGES.md.

Thanks to everyone who contributed to this release!

This release has several new features, many bug fixes, and numerous performance enhancements. To name a few:

• The new Ellipse3D sample consensus model
• The new farthest point sampling filter
• Faster and more robust PCD and PLY input/output

For an exhaustive list of newly added features, deprecations and other changes in PCL 1.13.0, please see CHANGES.md.

Thanks to everyone who contributed to this release!

This release has several new features, many bug fixes, and numerous performance enhancements. To name a few:

• The new Ellipse3D sample consensus model
• The new farthest point sampling filter
• Faster and more robust PCD and PLY input/output

For an exhaustive list of newly added features, deprecations and other changes
in PCL 1.13.0, please see CHANGES.md.

Thanks to everyone who contributed to this release!