Dataset: A video dataset of a wooden box assembly process

Jiayun Zhang; Petr Byvshev; Yu Xiao

doi:10.1145/3419016.3431492

Dataset: A video dataset of a wooden box assembly process

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

Abstract

This paper presents a video dataset of a 9-step wooden box assembly process including 17 subjects. The main strength of this dataset is the design of a standard and uniform workflow and the use of multiple cameras capturing videos from two different viewpoints. 62 video files were collected with the total size of 20 GB and the total duration of 13 hours. Each of the video is complemented with temporal annotations that indicate the starting and ending timestamps of each work step in the assembly process. We also provide statistical descriptive analyses of the recorded processes. Our dataset can be utilized for developing solutions to human activity recognition, process documentation, and many others that involve human-object interaction.

Original language	English
Title of host publication	DATA 2020 - Proceedings of the 3rd Workshop on Data Acquisition To Analysis, Part of SenSys 2020, BuildSys 2020
Publisher	ACM
Pages	35-39
Number of pages	5
ISBN (Electronic)	9781450381369
DOIs	https://doi.org/10.1145/3419016.3431492
Publication status	Published - 16 Nov 2020
MoE publication type	A4 Article in a conference publication
Event	ACM CONFERENCE ON EMBEDDED NETWORKED SENSOR SYSTEMS - Duration: 1 Jan 1900 → …

Conference

Conference	ACM CONFERENCE ON EMBEDDED NETWORKED SENSOR SYSTEMS
Abbreviated title	SenSys
Period	01/01/1900 → …

Keywords

Video Datasets
Assembly

Access to Document

10.1145/3419016.3431492

Dataset: A video dataset of a wooden box assembly processFinal published version, 2.04 MB

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Dataset: A video dataset of a wooden box assembly process'. Together they form a unique fingerprint.

2 Finished

DEBARE: Deep-learning Based Activity Recognition on the Edge
Xiao, Y., Parviainen, A., Pouta, E., Guridi Sotomayor, S. & Xie, J.
20/05/2019 → 31/10/2020
Project: EU: Framework programmes funding
CEAMA: Cognitive Engine for Assembly and Maintenance Automation
Lee, J., Pham, T., Wirtanen, S., Xiao, Y., Souza Leite, C., Byvshev, P., Pouta, E. & Nyman, P.
01/08/2018 → 31/01/2020
Project: Business Finland: New business from research ideas (TUTLI)

Cite this

@inproceedings{8e7fbcaf9cde48a8888de4d93113d995,

title = "Dataset: A video dataset of a wooden box assembly process",

abstract = "This paper presents a video dataset of a 9-step wooden box assembly process including 17 subjects. The main strength of this dataset is the design of a standard and uniform workflow and the use of multiple cameras capturing videos from two different viewpoints. 62 video files were collected with the total size of 20 GB and the total duration of 13 hours. Each of the video is complemented with temporal annotations that indicate the starting and ending timestamps of each work step in the assembly process. We also provide statistical descriptive analyses of the recorded processes. Our dataset can be utilized for developing solutions to human activity recognition, process documentation, and many others that involve human-object interaction.",

keywords = "Video Datasets, Assembly",

author = "Jiayun Zhang and Petr Byvshev and Yu Xiao",

year = "2020",

month = nov,

day = "16",

doi = "10.1145/3419016.3431492",

language = "English",

pages = "35--39",

booktitle = "DATA 2020 - Proceedings of the 3rd Workshop on Data Acquisition To Analysis, Part of SenSys 2020, BuildSys 2020",

publisher = "ACM",

note = "ACM CONFERENCE ON EMBEDDED NETWORKED SENSOR SYSTEMS, SenSys ; Conference date: 01-01-1900",

}

TY - GEN

T1 - Dataset: A video dataset of a wooden box assembly process

AU - Zhang, Jiayun

AU - Byvshev, Petr

AU - Xiao, Yu

PY - 2020/11/16

Y1 - 2020/11/16

N2 - This paper presents a video dataset of a 9-step wooden box assembly process including 17 subjects. The main strength of this dataset is the design of a standard and uniform workflow and the use of multiple cameras capturing videos from two different viewpoints. 62 video files were collected with the total size of 20 GB and the total duration of 13 hours. Each of the video is complemented with temporal annotations that indicate the starting and ending timestamps of each work step in the assembly process. We also provide statistical descriptive analyses of the recorded processes. Our dataset can be utilized for developing solutions to human activity recognition, process documentation, and many others that involve human-object interaction.

AB - This paper presents a video dataset of a 9-step wooden box assembly process including 17 subjects. The main strength of this dataset is the design of a standard and uniform workflow and the use of multiple cameras capturing videos from two different viewpoints. 62 video files were collected with the total size of 20 GB and the total duration of 13 hours. Each of the video is complemented with temporal annotations that indicate the starting and ending timestamps of each work step in the assembly process. We also provide statistical descriptive analyses of the recorded processes. Our dataset can be utilized for developing solutions to human activity recognition, process documentation, and many others that involve human-object interaction.

KW - Video Datasets

KW - Assembly

UR - http://www.scopus.com/inward/record.url?scp=85097356341&partnerID=8YFLogxK

U2 - 10.1145/3419016.3431492

DO - 10.1145/3419016.3431492

M3 - Conference contribution

SP - 35

EP - 39

BT - DATA 2020 - Proceedings of the 3rd Workshop on Data Acquisition To Analysis, Part of SenSys 2020, BuildSys 2020

PB - ACM

T2 - ACM CONFERENCE ON EMBEDDED NETWORKED SENSOR SYSTEMS

Y2 - 1 January 1900

ER -

Dataset: A video dataset of a wooden box assembly process

Abstract

Conference

Keywords

Access to Document

DEBARE: Deep-learning Based Activity Recognition on the Edge

CEAMA: Cognitive Engine for Assembly and Maintenance Automation

Cite this