Distinctive 3D local deep descriptors

Poiesi, Fabio; Boscaini, Davide

doi:10.1109/icpr48806.2021.9411978

Cited by 41 publications

(37 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The source code will be made publicly available at https://github.com/fabiopoiesi/gedi. This paper extends our earlier work [9] in many aspects. We modify the deep neural network architecture used in [9] in order to process patches at multiple scales.…”

Section: Introductionsupporting

confidence: 81%

“…This paper extends our earlier work [9] in many aspects. We modify the deep neural network architecture used in [9] in order to process patches at multiple scales. We introduce QNet to effectively achieve multi-scale processing, which brings two main benefits.…”

Section: Introductionsupporting

confidence: 81%

“…The latter firstly transforms points into suitable representations (e.g. projection on another space [10], [17] or rigid transformation [9]), then encodes these new representations into descriptors. Descriptors can in turn be handcrafted or based on…”

Section: Introductionmentioning

confidence: 99%

“…We refer to these descriptors as global descriptors because they encode the global context. Although global descriptors can be computed efficiently, they lack generalisation ability when trained with data captured in a context and used in other contexts [9], [10]. Differently, descriptors that encode local geometric information have a greater generalisation ability, although being computationally less efficient than their global counterpart [9], [10], [16].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Learning general and distinctive 3D local deep descriptors for point cloud registration

Poiesi,

Boscaini

2021

Preprint

Self Cite

View full text Add to dashboard Cite

An effective 3D descriptor should be invariant to different geometric transformations, such as scale and rotation, repeatable in the case of occlusions and clutter, and generalisable in different contexts when data is captured with different sensors. We present a simple but yet effective method to learn generalisable and distinctive 3D local descriptors that can be used to register point clouds captured in different contexts with different sensors. Point cloud patches are extracted, canonicalised with respect to their local reference frame, and encoded into scale and rotation-invariant compact descriptors by a point permutation-invariant deep neural network. Our descriptors can effectively generalise across sensor modalities from locally and randomly sampled points. We evaluate and compare our descriptors with alternative handcrafted and deep learning-based descriptors on several indoor and outdoor datasets reconstructed using both RGBD sensors and laser scanners. Our descriptors outperform most recent descriptors by a large margin in terms of generalisation, and become the state of the art also in benchmarks where training and testing are performed in the same scenarios.

show abstract

Section: Introductionsupporting

confidence: 81%

Section: Introductionsupporting

confidence: 81%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning general and distinctive 3D local deep descriptors for point cloud registration

Poiesi,

Boscaini

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Additionally, we will also perform this evaluation against DIP [34] descriptors, which are an example of local descriptors generated using modern Deep Learning techniques. Unlike handcrafted features, these Deep Learning based methods require a high number of images to train.…”

Section: Descriptor Evaluationmentioning

confidence: 99%

Shape binary patterns: an efficient local descriptor and keypoint detector for point clouds

Romero-González

García-Varea

Gómez

2022

Multimed Tools Appl

View full text Add to dashboard Cite

Many of the research problems in robot vision involve the detection of keypoints, areas with salient information in the input images and the generation of local descriptors, that encode relevant information for such keypoints. Computer vision solutions have recently relied on Deep Learning techniques, which make extensive use of the computational capabilities available. In autonomous robots, these capabilities are usually limited and, consequently, images cannot be processed adequately. For this reason, some robot vision tasks still benefit from a more classic approach based on keypoint detectors and local descriptors. In 2D images, the use of binary representations for visual tasks has shown that, with lower computational requirements, they can obtain a performance comparable to classic real-value techniques. However, these achievements have not been fully translated to 3D images, where research is mainly focused on real-value approaches. Thus, in this paper, we propose a keypoint detector and local descriptor based on 3D binary patterns. The experimentation demonstrates that our proposal is competitive against state-of-the-art techniques, while its processing can be performed more efficiently.

show abstract

GIPSO: Geometrically Informed Propagation for Online Adaptation in 3D LiDAR Segmentation

Saltori

Krivosheev

Lathuilière

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Distinctive 3D local deep descriptors

Cited by 41 publications

References 27 publications

Learning general and distinctive 3D local deep descriptors for point cloud registration

Learning general and distinctive 3D local deep descriptors for point cloud registration

Shape binary patterns: an efficient local descriptor and keypoint detector for point clouds

GIPSO: Geometrically Informed Propagation for Online Adaptation in 3D LiDAR Segmentation

Contact Info

Product

Resources

About