new JOSS paper structure

Author: ga58lar

docs/paper.md

@@ -24,16 +24,13 @@ bibliography: paper.bib
 
 # Summary
 
-PointCloudCrafter is a C++ command-line interface (CLI) toolkit for the processing, manipulation, and analysis of three-dimensional point cloud data. The software provides a collection of compiled executables for integration into automated research pipelines, enabling common point cloud operations without requiring custom code.
-
-The toolkit supports the conversion and processing of data from multiple acquisition and storage formats, including ROS2 bag recordings, binary files, plain-text files, and standard PLY and PCD files. A central design feature of PointCloudCrafter is its schema-agnostic handling of per-point attributes. Arbitrary scalar fields associated with individual points are preserved throughout conversion, transformation, filtering, and aggregation steps, ensuring that auxiliary metadata remains available for downstream analysis. Furthermore, it allows the accumulation of multiple point clouds based on their timestamps and transforms.
-
+PointCloudCrafter is a C++ command-line interface (CLI) toolkit for the processing, manipulation, and analysis of three-dimensional point cloud data. The software provides a collection of compiled executables for integration into automated research pipelines, enabling common point cloud operations without requiring custom code.  
+The toolkit supports the conversion and processing of data from multiple acquisition and storage formats, including ROS2 bag recordings, binary files, plain-text files, and standard PLY and PCD files. A central design feature of PointCloudCrafter is its schema-agnostic handling of per-point attributes. Arbitrary scalar fields associated with individual points are preserved throughout conversion, transformation, filtering, and aggregation steps, ensuring that auxiliary metadata remains available for downstream analysis. Furthermore, it allows the accumulation of multiple point clouds based on their timestamps and transforms.  
 By combining support for robotic middleware data with batch-oriented file processing, PointCloudCrafter facilitates reproducible workflows across a range of scientific and engineering applications that rely on large-scale three-dimensional point cloud data.
 
 # Statement of need
 
-Three-dimensional point cloud data is widely used in robotics, autonomous driving, remote sensing, and scientific data analysis. In practice, such data is commonly encountered in two distinct representations. During acquisition and experimentation, point clouds are often recorded as time-stamped data streams using robotic middleware such as ROS2. For benchmarking, archival, and offline analysis, the same data is typically stored and distributed as static files in dataset-specific or standardized formats. Especially, the use of automated pipelines for neural network training often requires the input data to be static files with a specific naming scheme.
-
+Three-dimensional point cloud data is widely used in robotics, autonomous driving, remote sensing, and scientific data analysis. In practice, such data is commonly encountered in two distinct representations. During acquisition and experimentation, point clouds are often recorded as time-stamped data streams using robotic middleware such as ROS2. For benchmarking, archival, and offline analysis, the same data is typically stored and distributed as static files in dataset-specific or standardized formats. Especially, the use of automated pipelines for neural network training often requires the input data to be static files with a specific naming scheme.  
 Transitioning between these representations, or performing large-scale batch operations on point cloud collections, remains a recurring challenge.
 
 The Point Cloud Library (PCL) [@Rusu2011] provides a comprehensive set of data structures and algorithms for three-dimensional processing and serves as the technical foundation of PointCloudCrafter, but its use generally requires direct integration into custom software projects. Graphical tools such as CloudCompare are well-suited for interactive inspection but are not designed for automated, headless processing of large datasets. Python-based libraries such as Open3D [@Zhou2018] and Pyoints [@Lamprecht2019] offer accessible interfaces for point cloud manipulation, but performance and input-output overhead can become limiting factors when processing large directories of high-resolution scans.
@@ -44,25 +41,43 @@ A further limitation of many existing tools is the implicit assumption of fixed-
 
 PointCloudCrafter addresses these needs by providing a lightweight CLI toolkit focused on batch-oriented point cloud conversion and processing. The primary software contribution is a schema-agnostic, command-line-driven framework that unifies middleware-based and file-based point cloud processing while preserving arbitrary per-point attributes. Arbitrary scalar fields refer to dynamically discovered per-point attributes whose names, data types, and memory layout are preserved without enforcing a predefined schema.
 
-# Functionality
+# Software design
+
+PointCloudCrafter is implemented in C++ and is designed primarily as a command-line toolkit composed of standalone executables for point cloud processing. The software builds on the Point Cloud Library (PCL) [@Rusu2011], which provides the underlying data structures and algorithms for ROS2 point cloud streaming and for commonly used file formats such as PCD. PointCloudCrafter does not reimplement PCL functionality; instead, it exposes selected operations and extends them through a unified, reproducible command-line interface.
+
+The software is developed using modern C++ design practices, including a clear separation between executable logic and shared processing components. This design improves maintainability and allows new command-line tools to be added in the future without altering existing workflows. Documentation of the codebase and available commands is generated with Doxygen and published as a static website on GitHub.
+
+The toolkit consists of two main executables that target distinct workflows. These executables share a common internal library that implements functionality for loading, saving, analyzing, transforming, filtering, and merging point clouds. Its purpose is to ensure consistent behavior and reduce code duplication across the executables, while keeping the user-facing interface focused on command-line usage.  
+Internally, all point cloud data is represented using the `PCLPointCloud2` data structure. This representation enables schema-agnostic handling of point clouds by allowing arbitrary per-point attributes to be discovered and preserved at runtime without enforcing a predefined schema.
 
 PointCloudCrafter provides two primary execution modes, *rosbag* and *file*, for streaming and static data workflows, respectively.
+The *rosbag* executable enables extraction of `sensor_msgs::msg::PointCloud2` topics from ROS2 bag recordings. This mode integrates with the `TF2` library [@Foote2013] to perform time-resolved pose lookups based on the recorded transform tree. Each point cloud can be transformed from the sensor coordinate frame into a user-specified fixed reference frame using the pose corresponding to the exact acquisition timestamp. This enables motion-consistent export of point cloud data from dynamic sensor platforms and a timestamp-based fusion of different point cloud sensor topics.  
+The *file* executable focuses on batch-oriented processing of static point cloud files. Supported formats include common text and binary formats such as PLY, PCD, OBJ, and TXT, as well as dataset-specific binary formats used in KITTI [@KITTI2012] and nuScenes [@nuScenes2019].  
+Both executables support a common set of processing operations implemented in the shared internal codebase. It allows users to apply rigid-body transformations, merge multiple point clouds into a single representation, analyse timestamps, and perform filtering operations. Filters include geometric cropping using box, sphere, or cylindrical regions, voxel grid downsampling, and statistical or radius-based outlier removal. All processing steps are fully parameterized via command-line options, enabling deterministic and reproducible execution across datasets. Due to the schema-agnostic internal representation, per-point attributes, such as sensor-specific metadata or domain-specific properties, are preserved throughout processing whenever supported by the input and output formats.
+
+A comprehensive overview of the available commands, parameters, and usage examples is provided in the online documentation at \url{https://tumftm.github.io/PointCloudCrafter/usage.html}.
+
+# Research impact statement
 
-The *rosbag* mode enables extraction of `sensor_msgs::msg::PointCloud2` topics from ROS2 bag recordings. This mode integrates with the `TF2` library [@Foote2013] to perform time-resolved pose lookups based on the recorded transform tree. Each point cloud can be transformed from the sensor coordinate frame into a user-specified fixed reference frame using the pose corresponding to the exact acquisition timestamp. This enables motion-consistent export of point cloud data from dynamic sensor platforms and a timestamp-based fusion of different point cloud sensor topics.
+PointCloudCrafter has been developed to support reproducible preprocessing of large-scale point cloud datasets in research environments. Early versions of the toolkit have been used in student research projects, scientific publications [@Leitenstern2025; @Kulmer2025-1], and the creation of publicly available datasets [@Kulmer2025-2].
+It has been used in a range of internal research workflows related to autonomous driving and robotic perception. These applications include analyzing sensor drivers by inspecting timestamp consistency and packet loss, which manifest as incomplete or irregular point clouds, as well as fusing multiple ROS2 LiDAR topics into a unified point cloud representation using a user-specified static reference frame. The toolkit has been further applied to the creation of custom datasets for evaluating state-of-the-art open-source algorithms in pseudo-labeling, object detection, localization, and mapping. In addition, PointCloudCrafter has been integrated into automated data processing pipelines that convert logged ROS2 data into established dataset formats, such as KITTI, to enable systematic evaluation and labeling for subsequent machine learning training. The software has also been used to generate large-scale, high-resolution point cloud maps by aggregating temporally indexed scans using externally estimated sensor poses.
 
-The *file* mode focuses on batch processing of static point cloud files. Supported formats include common text and binary formats such as PLY, PCD, OBJ, and TXT, as well as dataset-specific binary formats used in KITTI [@KITTI2012] and nuScenes [@nuScenes2019]. 
+By providing a middleware-aware, schema-agnostic CLI toolkit, PointCloudCrafter lowers the barrier for converting raw sensor data into analysis-ready datasets. Its focus on ease of use, reproducibility, and preservation of per-point metadata makes it applicable across multiple domains that rely on three-dimensional point cloud data.
 
-Both execution modes rely internally on the `PCLPointCloud2` data structure, enabling schema-agnostic handling of point attributes. They allow users to apply rigid-body transformations, merge multiple point clouds into a single representation, and perform filtering operations. Implemented filters include geometric cropping using box, sphere, or cylindrical regions, voxel grid downsampling, and statistical or radius-based outlier removal. As a result of the implemented data formats, the processed point clouds can also be converted into other data formats.
-All processing steps are fully parameterized via command-line options, enabling deterministic, reproducible execution across datasets.
+# Availability
 
-A comprehensive overview of the functions and the parameterization can be found at \url{https://tumftm.github.io/PointCloudCrafter/usage.html}
+The source code is available at \url{https://github.com/TUMFTM/PointCloudCrafter}.  
+The Docker is available at \url{https://ghcr.io/tumftm/pointcloudcrafter:latest}.  
+The documentation is available at \url{https://tumftm.github.io/PointCloudCrafter}.  
+The license is Apache-2.0.  
 
-A typical use case is the extraction of LiDAR scans from ROS2 bag recordings, transformation into a global reference frame using an externally estimated trajectory, and aggregation into a static map for offline evaluation. This workflow can be executed in a single batch process while preserving sensor-specific scalar fields required for subsequent analysis.
+# AI usage disclosure
 
-Early versions of PointCloudCrafter have been used in diverse student research projects, scientific publications [@Leitenstern2025; @Kulmer2025-1], and the creation of datasets [@Kulmer2025-2].
+Artificial intelligence tools were used in a limited and transparent manner during the development of this work. Claude Opus 4.5 was used to assist with code structuring and to generate code comments. GitHub Copilot was partially used to review Pull Request. The manuscript text was reviewed for grammar and clarity using OpenAI GPT-5.2 and Grammarly, and selected passages were rephrased to improve readability and flow. All scientific content, software design decisions, and final wording were reviewed and approved by the authors, who retain full responsibility for the work.
 
 # Acknowledgements
 
-The authors acknowledge the developers and maintainers of the Point Cloud Library (PCL) [@Rusu2011], upon which PointCloudCrafter is built. We also acknowledge the contributions of student researchers at the Institute of Automotive Technology at the Technical University of Munich who supported the development and evaluation of this software.
+The authors acknowledge the developers and maintainers of the Point Cloud Library (PCL) [@Rusu2011], upon which PointCloudCrafter is built. We also acknowledge the contributions of student researchers at the Institute of Automotive Technology at the Technical University of Munich who supported the development and evaluation of this software.  
+This work was funded by basic research funds of the Institute of Automotive Technology at the Technical University of Munich.
 
 # References