Motion pattern recognition in 4D point clouds

Dariush Salami; Sameera Palipana; Manila Kodali; Stephan Sigg

doi:10.1109/MLSP49062.2020.9231569

Motion pattern recognition in 4D point clouds

Dariush Salami, Sameera Palipana, Manila Kodali, Stephan Sigg

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

6 Citations (Scopus)

141 Downloads (Pure)

Abstract

We address an actively discussed problem in signal processing, recognizing patterns from spatial data in motion. In particular, we suggest a neural network architecture to recognize motion patterns from 4D point clouds. We demonstrate the feasibility of our approach with point cloud datasets of hand gestures. The architecture, PointGest, directly feeds on unprocessed timelines of point cloud data without any need for voxelization or projection. The model is resilient to noise in the input point cloud through abstraction to lower-density representations, especially for regions of high density. We evaluate the architecture on a benchmark dataset with ten gestures. PointGest achieves an accuracy of 98.8%, outperforming five state-of-the-art point cloud classification models.

Original language	English
Title of host publication	Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781728166629
DOIs	https://doi.org/10.1109/MLSP49062.2020.9231569
Publication status	Published - Sep 2020
MoE publication type	A4 Article in a conference publication
Event	IEEE International Workshop on Machine Learning for Signal Processing - Aalto University, Espoo, Finland Duration: 21 Sep 2020 → 24 Sep 2020 Conference number: 30 https://ieeemlsp.cc

Publication series

Name	IEEE International Workshop on Machine Learning for Signal Processing
ISSN (Print)	2161-0363
ISSN (Electronic)	2161-0371

Workshop

Workshop	IEEE International Workshop on Machine Learning for Signal Processing
Abbreviated title	MLSP
Country/Territory	Finland
City	Espoo
Period	21/09/2020 → 24/09/2020
Internet address	https://ieeemlsp.cc

Keywords

4D point clouds
Deep learning
Gesture recognition

Access to Document

10.1109/MLSP49062.2020.9231569

Salami_Motion_pattern_recognition_in_4DAccepted author manuscript, 492 KB

OpenUrl availability

Full text

Cite this

@inproceedings{a642ae68b5084dfab5a0795be87b045e,

title = "Motion pattern recognition in 4D point clouds",

abstract = "We address an actively discussed problem in signal processing, recognizing patterns from spatial data in motion. In particular, we suggest a neural network architecture to recognize motion patterns from 4D point clouds. We demonstrate the feasibility of our approach with point cloud datasets of hand gestures. The architecture, PointGest, directly feeds on unprocessed timelines of point cloud data without any need for voxelization or projection. The model is resilient to noise in the input point cloud through abstraction to lower-density representations, especially for regions of high density. We evaluate the architecture on a benchmark dataset with ten gestures. PointGest achieves an accuracy of 98.8%, outperforming five state-of-the-art point cloud classification models. ",

keywords = "4D point clouds, Deep learning, Gesture recognition",

author = "Dariush Salami and Sameera Palipana and Manila Kodali and Stephan Sigg",

note = "| openaire: EC/H2020/813999/EU//WINDMILL; IEEE International Workshop on Machine Learning for Signal Processing, MLSP ; Conference date: 21-09-2020 Through 24-09-2020",

year = "2020",

month = sep,

doi = "10.1109/MLSP49062.2020.9231569",

language = "English",

series = "IEEE International Workshop on Machine Learning for Signal Processing",

publisher = "IEEE",

booktitle = "Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020",

address = "United States",

url = "https://ieeemlsp.cc",

}

Salami, D, Palipana, S, Kodali, M & Sigg, S 2020, Motion pattern recognition in 4D point clouds. in Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020., 9231569, IEEE International Workshop on Machine Learning for Signal Processing, IEEE, IEEE International Workshop on Machine Learning for Signal Processing, Espoo, Finland, 21/09/2020. https://doi.org/10.1109/MLSP49062.2020.9231569

Motion pattern recognition in 4D point clouds. / Salami, Dariush; Palipana, Sameera; Kodali, Manila et al.

Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020. IEEE, 2020. 9231569 (IEEE International Workshop on Machine Learning for Signal Processing).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Motion pattern recognition in 4D point clouds

AU - Salami, Dariush

AU - Palipana, Sameera

AU - Kodali, Manila

AU - Sigg, Stephan

N1 - Conference code: 30

PY - 2020/9

Y1 - 2020/9

N2 - We address an actively discussed problem in signal processing, recognizing patterns from spatial data in motion. In particular, we suggest a neural network architecture to recognize motion patterns from 4D point clouds. We demonstrate the feasibility of our approach with point cloud datasets of hand gestures. The architecture, PointGest, directly feeds on unprocessed timelines of point cloud data without any need for voxelization or projection. The model is resilient to noise in the input point cloud through abstraction to lower-density representations, especially for regions of high density. We evaluate the architecture on a benchmark dataset with ten gestures. PointGest achieves an accuracy of 98.8%, outperforming five state-of-the-art point cloud classification models.

AB - We address an actively discussed problem in signal processing, recognizing patterns from spatial data in motion. In particular, we suggest a neural network architecture to recognize motion patterns from 4D point clouds. We demonstrate the feasibility of our approach with point cloud datasets of hand gestures. The architecture, PointGest, directly feeds on unprocessed timelines of point cloud data without any need for voxelization or projection. The model is resilient to noise in the input point cloud through abstraction to lower-density representations, especially for regions of high density. We evaluate the architecture on a benchmark dataset with ten gestures. PointGest achieves an accuracy of 98.8%, outperforming five state-of-the-art point cloud classification models.

KW - 4D point clouds

KW - Deep learning

KW - Gesture recognition

UR - http://www.scopus.com/inward/record.url?scp=85096462752&partnerID=8YFLogxK

U2 - 10.1109/MLSP49062.2020.9231569

DO - 10.1109/MLSP49062.2020.9231569

M3 - Conference contribution

AN - SCOPUS:85096462752

T3 - IEEE International Workshop on Machine Learning for Signal Processing

BT - Proceedings of the 2020 IEEE 30th International Workshop on Machine Learning for Signal Processing, MLSP 2020

PB - IEEE

T2 - IEEE International Workshop on Machine Learning for Signal Processing

Y2 - 21 September 2020 through 24 September 2020

ER -

Motion pattern recognition in 4D point clouds

Abstract

Publication series

Workshop

Keywords

Access to Document

OpenUrl availability

Other files and links

Fingerprint

WINDMILL: Integrating Wireless Communication ENgineering and MachIne Learning

RadioSense: Wireless Big Data Augmented Smart Industry

Science-IT

Cite this

Motion pattern recognition in 4D point clouds

Abstract

Publication series

Workshop

Keywords

Access to Document

OpenUrl availability

Other files and links

Fingerprint

Projects

WINDMILL: Integrating Wireless Communication ENgineering and MachIne Learning

RadioSense: Wireless Big Data Augmented Smart Industry

Equipment

Science-IT

Cite this