NeeDrop: Self-supervised Shape Representation from Sparse Point Clouds using Needle Dropping

Boulch, Alexandre; Langlois, Pierre-Alain; Puy, Gilles; Marlet, Renaud

doi:10.1109/3dv53792.2021.00102

Cited by 14 publications

(11 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Needrop [11] is not an overfitting method; it uses global shape latent codes to learn to reconstruct shapes as occupancy field from sparse, noisy, and normal-less point clouds. It uses an unique approach inspired by Buffon's needle problem.…”

Section: Overfit a Single Shapementioning

confidence: 99%

A Review of Deep Learning-Powered Mesh Reconstruction Methods

Feng¹

2023

Preprint

View full text Add to dashboard Cite

With the recent advances in hardware and rendering techniques, 3D models have emerged everywhere in our life. Yet creating 3D shapes is arduous and requires significant professional knowledge. Meanwhile, Deep learning has enabled high-quality 3D shape reconstruction from various sources, making it a viable approach to acquiring 3D shapes with minimal effort. Importantly, to be used in common 3D applications, the reconstructed shapes need to be represented as polygonal meshes, which is a challenge for neural networks due to the irregularity of mesh tessellations.In this survey, we provide a comprehensive review of mesh reconstruction methods that are powered by machine learning. We first describe various representations for 3D shapes in the deep learning context. Then we review the development of 3D mesh reconstruction methods from voxels, point clouds, single images, and multi-view images. Finally, we identify several challenges in this field and propose potential future directions.

show abstract

Section: Overfit a Single Shapementioning

confidence: 99%

A Review of Deep Learning-Powered Mesh Reconstruction Methods

Feng¹

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…2) Surface information: Except a few unsupervised methods that can learn surface information from raw points clouds [54], most learning-based reconstruction methods rely directly or indirectly on supervision from mesh datasets, synthetic (CAD-based) like ShapeNet [18] or SceneNet [19], or real (based on actual scans) like DFaust [55] or MatterPort3D [53].…”

Section: B Datasetsmentioning

confidence: 99%

VASAD: a Volume and Semantic dataset for Building Reconstruction from Point Clouds

Langlois¹,

Xiao²,

Boulch

et al. 2022

2022 26th International Conference on Pattern Recognition (ICPR)

Self Cite

View full text Add to dashboard Cite

3D scene reconstruction has important applications to help to produce digital twins of existing buildings. While the community has mostly focused on surface reconstruction or semantic segmentation as separate problems, the joint reconstruction of both volumes and semantics has little been discussed, mostly due to the lack of large scale volume datasets with semantic annotations. In this work, we introduce a new dataset called VASAD for Volume And Semantic Architectural Dataset. It is composed of 6 building models, with full volume description and semantic labels. It approximately represents 62,000 m 2 of building floors, making it large enough for the development and evaluation of learning-based approaches. We propose several methods to jointly reconstruct both geometry and semantics and evaluate on the test set of the dataset. We show that the proposed dataset is challenging enough to stimulate research. The dataset is available at https://github.com/palanglois/vasad.

show abstract

“…Kundu [13] uses meta-learning to find a good category-specific initialization and employ instance-specific fully weight encoded functions to represent each object in scene. Most of the previous work is based on dense point clouds or images to complete instance reconstruction, Boulch [6] adopted sampling strategy of picking needles with end points on opposite sides or on the same side of the surface to realize dense reconstruction with sparse point cloud, this method requires pre-training on the dataset where sparse point clouds reside.…”

Section: Related Workmentioning

confidence: 99%

City-scale Incremental Neural Mapping with Three-layer Sampling and Panoptic Representation

Shi¹,

Yang²,

Li³

et al. 2022

Preprint

View full text Add to dashboard Cite

Neural implicit representations are drawing a lot of attention from the robotics community recently, as they are expressive, continuous and compact. However, cityscale incremental implicit dense mapping based on sparse LiDAR input is still an under-explored challenge. To this end, we successfully build the first city-scale incremental neural mapping system with a panoptic representation that consists of both environment-level and instance-level modelling. Given a stream of sparse LiDAR point cloud, it maintains a dynamic generative model that maps 3D coordinates to signed distance field (SDF) values. To address the difficulty of representing geometric information at different levels in city-scale space, we propose a tailored three-layer sampling strategy to dynamically sample the global, local and near-surface domains. Meanwhile, to realize high fidelity mapping, category-specific prior is introduced to better model the geometric details, leading to a panoptic representation. We evaluate on the public SemanticKITTI dataset and demonstrate the significance of the newly proposed three-layer sampling strategy and panoptic representation, using both quantitative and qualitative results. Codes and data will be publicly available.

show abstract

NeeDrop: Self-supervised Shape Representation from Sparse Point Clouds using Needle Dropping

Cited by 14 publications

References 41 publications

A Review of Deep Learning-Powered Mesh Reconstruction Methods

A Review of Deep Learning-Powered Mesh Reconstruction Methods

VASAD: a Volume and Semantic dataset for Building Reconstruction from Point Clouds

City-scale Incremental Neural Mapping with Three-layer Sampling and Panoptic Representation

Contact Info

Product

Resources

About