Papers
arxiv:2604.10780

LIDARLearn: A Unified Deep Learning Library for 3D Point Cloud Classification, Segmentation, and Self-Supervised Representation Learning

Published on Apr 12
Authors:
,
,

Abstract

A unified PyTorch library for 3D point cloud analysis that standardizes model implementations, training protocols, and evaluation methods across multiple deep learning approaches.

AI-generated summary

Three-dimensional (3D) point cloud analysis has become central to applications ranging from autonomous driving and robotics to forestry and ecological monitoring. Although numerous deep learning methods have been proposed for point cloud understanding, including supervised backbones, self-supervised pre-training (SSL), and parameter-efficient fine-tuning (PEFT), their implementations are scattered across incompatible codebases with differing data pipelines, evaluation protocols, and configuration formats, making fair comparisons difficult. We introduce , a unified, extensible PyTorch library that integrates over 55 model configurations covering 29 supervised architectures, seven SSL pre-training methods, and five PEFT strategies, all within a single registry-based framework supporting classification, semantic segmentation, part segmentation, and few-shot learning. provides standardised training runners, cross-validation with stratified K-fold splitting, automated LaTeX/CSV table generation, built-in Friedman/Nemenyi statistical testing with critical-difference diagrams for rigorous multi-model comparison, and a comprehensive test suite with 2\,200+ automated tests validating every configuration end-to-end. The code is available at https://github.com/said-ohamouddou/LIDARLearn under the MIT licence.

Community

Paper author

We’re proud to open-source LIDARLearn 🎉

It’s a unified PyTorch library for 3D point cloud deep learning. To our knowledge, it’s the first framework that supports such a large collection of models in one place, with built-in cross-validation support.

It brings together 56 ready-to-use configurations covering supervised, self-supervised, and parameter-efficient fine-tuning methods.

You can run everything from a single YAML file with one simple command.

One of the best features: after training, you can automatically generate a publication-ready LaTeX PDF. It creates clean tables, highlights the best results, and runs statistical tests and diagrams for you. No need to build tables manually in Overleaf.

The library includes benchmarks on datasets like ModelNet40, ShapeNet, S3DIS, and two remote sensing datasets (STPCTLS and HELIALS). STPCTLS is already preprocessed, so you can use it right away.

This project is intended for researchers in 3D point cloud learning, 3D computer vision, and remote sensing.

It’s released under the MIT license.

Contributions and benchmarks are welcome!

GitHub 💻: https://github.com/said-ohamouddou/LIDARLearn

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2604.10780
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2604.10780 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2604.10780 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2604.10780 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.