BUP-ST20: Weakly Labelled Spatial Temporal Sweet Pepper Data

Name: BUP-ST20: Weakly Labelled Spatial Temporal Sweet Pepper Data
Published: 2025-10-06
License: http://creativecommons.org/licenses/by/4.0

Open this dataset in the live repository

Important usage condition: Use and redistribution are governed by CC BY 4.0. Review and comply with the license before using the data.

Persistent identifier: doi:10.60507/FK2/NUMVO1
Published version: 2.0
Publication date: 2025-10-06
License: CC BY 4.0

Description

Accurate monitoring of crop phenotypic traits is essential for efficient farm management and automation in agriculture. Multi-object tracking (MOT) and video instance segmentation (VIS) offer promising approaches to enhance agricultural robotic vision systems, yet a major limitation is the scarcity of high-quality spatial-temporal datasets. We introduce BUP-ST20, a novel weakly labelled spatial-temporal dataset for sweet pepper tracking and segmentation captured on a robotic platform. BUP-ST20 contains 16,240 images from 275 sequences, each with bounding boxes, instance segmentation masks, and temporal identities.The dataset has weakly labelled training and validation sets, while the evaluation set includes 3810 frames with hand-labelled ground truth annotations.

Creators

Guclu, Esra
Halstead, Michael
Denman, Simon
McCool, Chris

Keywords

Multi-object tracking, video instance segmentation, spatial-temporal dataset

Files

Before downloading: Use and redistribution are governed by CC BY 4.0. Review and comply with the license before using the data.

File	Type	Bytes	Checksum
ReadMe.txt	text/plain	5647	MD5 9e1281c84ef1f5824d196d1c8ec426a2
bupst20_rgb.tar.gz.aa	audio/x-pn-audibleaudio	4294967296	MD5 ddc88c6f9dc72f1cbe6b534ec782d64c
bupst20_rgb.tar.gz.ae	application/octet-stream	4294967296	MD5 74b70f880f50c033bf2971013079144a
bupst20_depth.tar.gz.ab	application/octet-stream	910408094	MD5 a6b7214d30e3dd9b38d964a58fe71870
cam_params.yaml	application/x-yaml	280	MD5 ed212b2055b08b635f263320a42dec15
bupst20_annotations.tar.gz	application/gzip	201284699	MD5 006394bef43394378798f2c49304a9a8
bupst20_odometry.tar.gz	application/gzip	252713	MD5 2ac75afc241e63577b6f02c9efcd4de4
bupst20_depth.tar.gz.aa	audio/x-pn-audibleaudio	4294967296	MD5 e02acce17d6b766c3b70f539541a23a0
bupst20_rgb.tar.gz.ab	application/octet-stream	4294967296	MD5 81bc52858c4892402a1fedfe1ab90707
bupst20_rgb.tar.gz.ac	application/pkix-attr-cert	4294967296	MD5 93338ef9591e503e1dec8da6b604e2f9
bupst20_rgb.tar.gz.ag	image/x-applix-graphics	3779149996	MD5 abd54e3866adab851cb575d1d3a58d36
how_to_use_BUPST20.md	text/markdown	5206	MD5 30831e1ba9bab6fbaa91fd347ce5d855
dataset_structure.md	text/markdown	3009	MD5 aa1af827322e4ae5dc3f0dc668af7562
train_valid_eval_splits.yaml	application/x-yaml	1670	MD5 dcf653a6d147f9c6eb577b7f65ce000a
bupst20_rgb.tar.gz.af	application/octet-stream	4294967296	MD5 252b920bac08bc0930a21a770569a738
bupst20_rgb.tar.gz.ad	application/octet-stream	4294967296	MD5 6e19dd2a7a0444d63ee08e473a84301d

Citation

Guclu, Esra; Halstead, Michael; Denman, Simon; McCool, Chris, 2025-10-06, BUP-ST20: Weakly Labelled Spatial Temporal Sweet Pepper Data, doi:10.60507/FK2/NUMVO1, V2.0

Additional Dataverse fields

Id	485
Dataset Type	dataset
Internal Version Number	10
Latest Version Publishing State	RELEASED
Distribution Date	2025-09-29
Release Time	2025-10-13T09:54:43Z
Create Time	2025-10-07T04:15:24Z
Citation Date	2025-10-06
File Access Request	True

Export metadata

Static metadata exports available for this published dataset version:

Complete Dataverse metadata

Citation Metadata

Title title

BUP-ST20: Weakly Labelled Spatial Temporal Sweet Pepper Data

Author author

Author Name
Guclu, Esra
Author Affiliation
University of Bonn
Author Identifier Scheme
ORCID
Author Identifier
Type Name
authorIdentifier
Multiple
False
Type Class
primitive
Value
https://orcid.org/0000-0003-1106-5566
Expandedvalue
Person Name
GUCLU, ESRA
@id
https://orcid.org/0000-0003-1106-5566
Scheme
ORCID
@type
https://schema.org/Person
Author Name
Halstead, Michael
Author Affiliation
University of Bonn
Author Identifier Scheme
ORCID
Author Identifier
Type Name
authorIdentifier
Multiple
False
Type Class
primitive
Value
https://orcid.org/0000-0001-7185-9304
Expandedvalue
Person Name
Halstead, Michael
@id
https://orcid.org/0000-0001-7185-9304
Scheme
ORCID
@type
https://schema.org/Person
Author Name
Denman, Simon
Author Affiliation
Queensland University of Technology
Author Identifier Scheme
ORCID
Author Identifier
Type Name
authorIdentifier
Multiple
False
Type Class
primitive
Value
https://orcid.org/0000-0002-0983-5480
Expandedvalue
Person Name
Denman, Simon
@id
https://orcid.org/0000-0002-0983-5480
Scheme
ORCID
@type
https://schema.org/Person
Author Name
McCool, Chris
Author Affiliation
University of Bonn
Author Identifier Scheme
ORCID
Author Identifier
Type Name
authorIdentifier
Multiple
False
Type Class
primitive
Value
https://orcid.org/0000-0002-0577-1299
Expandedvalue
Person Name
McCool, Chris
@id
https://orcid.org/0000-0002-0577-1299
Scheme
ORCID
@type
https://schema.org/Person

Dataset Contact datasetContact

Dataset Contact Name
Güclü, Esra
Dataset Contact Affiliation
University of Bonn
Dataset Contact Email
egueclue@uni-bonn.de

Ds Description dsDescription

Ds Description Value
Accurate monitoring of crop phenotypic traits is essential for efficient farm management and automation in agriculture. Multi-object tracking (MOT) and video instance segmentation (VIS) offer promising approaches to enhance agricultural robotic vision systems, yet a major limitation is the scarcity of high-quality spatial-temporal datasets. We introduce BUP-ST20, a novel weakly labelled spatial-temporal dataset for sweet pepper tracking and segmentation captured on a robotic platform. BUP-ST20 contains 16,240 images from 275 sequences, each with bounding boxes, instance segmentation masks, and temporal identities.The dataset has weakly labelled training and validation sets, while the evaluation set includes 3810 frames with hand-labelled ground truth annotations.

Subject subject

Agricultural Sciences

Free Keyword freeKeyword

Free Keyword Value
Multi-object tracking
Free Keyword Value
video instance segmentation
Free Keyword Value
spatial-temporal dataset

Publication publication

Publication Relation Type
IsSupplementTo
Publication Citation
Guclu E, Halstead M, Denman S, McCool C. Weakly labelled spatial-temporal sweet pepper data: Enabling higher quality detection, segmentation, and tracking. The International Journal of Robotics Research. 2025;0(0). doi:10.1177/02783649251379093
Publication IDType
doi
Publication IDNumber
10.1177/02783649251379093
Publication URL
https://doi.org/10.1177/02783649251379093

Language language

English

Contributor contributor

Contributor Type
Research Group
Contributor Name
Agricultural Robotics Group
Contributor Type
Data Curator
Contributor Name
Guclu Esra
Contributor Type
Researcher
Contributor Name
Halstead Michael
Contributor Type
Supervisor
Contributor Name
McCool Chris
Contributor Type
Researcher
Contributor Name
Denman Simon

Distribution Date distributionDate

2025-09-29

Depositor depositor

Güclü, Esra

Date Of Deposit dateOfDeposit

2025-09-29

Date Of Collection dateOfCollection

Date Of Collection Start
2020-09-24
Date Of Collection End
2020-10-01

Kind Of Data kindOfData

Images
Numerical
Spreadsheet
Structured
Qualitative

Computational Workflow Metadata

Workflow Type workflowType

Bash Script
Other Python-based workflow

Expected crawler behaviour

Use a stable, truthful User-Agent with product/version and a working contact URL. Across all IP addresses and HTTP connections used by one crawler identity, allow no more than 5 requests in flight and wait at least 20 seconds between request starts. Crawl URLs listed in the catalog sitemap, including file pages and download URLs when they are published, use conditional requests, honor Retry-After, and apply exponential backoff after errors.

The welcome page may link to the interactive repository for human navigation. Automated clients must not treat that human link as a catalog crawl target.

Read the live machine-readable crawler policy before and during a crawl. Stop crawling when it reports CPU or memory utilization at or above 80% and 80% respectively.