# SPU-Net: Self-Supervised Point Cloud Upsampling by Coarse-to-Fine Reconstruction with Self-Projection Optimization

arXiv: https://arxiv.org/abs/2012.04439 · HTML: https://mlsanity.com/paper/2012.04439/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2020.

## Built on by (most cited first, 26 of 26)

- [CAP-UDF: Learning Unsigned Distance Functions Progressively from Raw Point Clouds with Consistency-Aware Field Optimization](https://mlsanity.com/paper/2210.02757/) (2022)
- [Learning a More Continuous Zero Level Set in Unsigned Distance Fields through Level Set Projection](https://mlsanity.com/paper/2308.11441/) (2023)
- [PU-Transformer: Point Cloud Upsampling Transformer](https://mlsanity.com/paper/2111.12242/) (2021)
- [Self-Supervised Arbitrary-Scale Point Clouds Upsampling via Implicit Neural Representation](https://mlsanity.com/paper/2204.08196/) (2022)
- [Learning Continuous Implicit Field with Local Distance Indicator for Arbitrary-Scale Point Cloud Upsampling](https://mlsanity.com/paper/2312.15133/) (2023)
- [3D-OAE: Occlusion Auto-Encoders for Self-Supervised Learning on Point Clouds](https://mlsanity.com/paper/2203.14084/) (2022)
- [PC2-PU: Patch Correlation and Point Correlation for Effective Point Cloud Upsampling](https://mlsanity.com/paper/2109.09337/) (2021)
- [GQE-Net: A Graph-based Quality Enhancement Network for Point Cloud Color Attribute](https://mlsanity.com/paper/2303.13764/) (2023)
- [Multiview Textured Mesh Recovery by Differentiable Rendering](https://mlsanity.com/paper/2205.12468/) (2022)
- [Attention Models for Point Clouds in Deep Learning: A Survey](https://mlsanity.com/paper/2102.10788/) (2021)
- [Retro-FPN: Retrospective Feature Pyramid Network for Point Cloud Semantic Segmentation](https://mlsanity.com/paper/2308.09314/) (2023)
- [Hierarchical Prior-based Super Resolution for Point Cloud Geometry Compression](https://mlsanity.com/paper/2402.11250/) (2024)
- [Deep Learning for 3D Point Cloud Enhancement: A Survey](https://mlsanity.com/paper/2411.00857/) (2024)
- [Self-Ordering Point Clouds](https://mlsanity.com/paper/2304.00961/) (2023)
- [Parametric Surface Constrained Upsampler Network for Point Cloud](https://mlsanity.com/paper/2303.08240/) (2023)
- [PU-Ray: Domain-Independent Point Cloud Upsampling via Ray Marching on Neural Implicit Surface](https://mlsanity.com/paper/2310.08755/) (2023)
- [Herd Mentality in Augmentation -- Not a Good Idea! A Robust Multi-stage Approach towards Deepfake Detection](https://mlsanity.com/paper/2410.05466/) (2024)
- [SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network](https://mlsanity.com/paper/2502.19452/) (2025)
- [Learning Structural Latent Points for Efficient Visual Representations in Robotic Manipulation](https://mlsanity.com/paper/2605.21258/) (2026)
- [GoogLe2Net: Going Transverse with Convolutions](https://mlsanity.com/paper/2301.00424/) (2023)
- [Self-Supervised Learning for Point Clouds Data: A Survey](https://mlsanity.com/paper/2305.11881/) (2023)
- [Point Cloud Novelty Detection Based on Latent Representations of a General Feature Extractor](https://mlsanity.com/paper/2410.09861/) (2024)
- [Methods for evaluating the resolution of 3D data derived from satellite images](https://mlsanity.com/paper/2506.11876/) (2025)
- [PVNet: Point-Voxel Interaction LiDAR Scene Upsampling Via Diffusion Models](https://mlsanity.com/paper/2508.17050/) (2025)
- [PUFM++: Point Cloud Upsampling via Enhanced Flow Matching](https://mlsanity.com/paper/2512.20988/) (2025)
- [Point Cloud Upsampling through Patch-based Frequency Superposition](https://mlsanity.com/paper/2606.14355/) (2026)

## Abstract (arXiv)

The task of point cloud upsampling aims to acquire dense and uniform point sets from sparse and irregular point sets. Although significant progress has been made with deep learning models, state-of-the-art methods require ground-truth dense point sets as the supervision, which makes them limited to be trained under synthetic paired training data and not suitable to be under real-scanned sparse data. However, it is expensive and tedious to obtain large numbers of paired sparse-dense point sets as supervision from real-scanned sparse data. To address this problem, we propose a self-supervised point cloud upsampling network, named SPU-Net, to capture the inherent upsampling patterns of points lying on the underlying object surface. Specifically, we propose a coarse-to-fine reconstruction framework, which contains two main components: point feature extraction and point feature expansion, respectively. In the point feature extraction, we integrate the self-attention module with the graph convolution network (GCN) to capture context information inside and among local regions simultaneously. In the point feature expansion, we introduce a hierarchically learnable folding strategy to generate upsampled point sets with learnable 2D grids. Moreover, to further optimize the noisy points in the generated point sets, we propose a novel self-projection optimization associated with uniform and reconstruction terms as a joint loss to facilitate the self-supervised point cloud upsampling. We conduct various experiments on both synthetic and real-scanned datasets, and the results demonstrate that we achieve comparable performances to state-of-the-art supervised methods.