Not Every Patch is Needed: Towards a More Efficient and Effective Backbone for Video-based Person Re-identification Article Swipe

PDF

Lanyun Zhu , Tianrun Chen , Deyi Ji , Jieping Ye , Jun Liu ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2501.16811

This paper proposes a new effective and efficient plug-and-play backbone for video-based person re-identification (ReID). Conventional video-based ReID methods typically use CNN or transformer backbones to extract deep features for every position in every sampled video frame. Here, we argue that this exhaustive feature extraction could be unnecessary, since we find that different frames in a ReID video often exhibit small differences and contain many similar regions due to the relatively slight movements of human beings. Inspired by this, a more selective, efficient paradigm is explored in this paper. Specifically, we introduce a patch selection mechanism to reduce computational cost by choosing only the crucial and non-repetitive patches for feature extraction. Additionally, we present a novel network structure that generates and utilizes pseudo frame global context to address the issue of incomplete views resulting from sparse inputs. By incorporating these new designs, our backbone can achieve both high performance and low computational cost. Extensive experiments on multiple datasets show that our approach reduces the computational cost by 74\% compared to ViT-B and 28\% compared to ResNet50, while the accuracy is on par with ViT-B and outperforms ResNet50 significantly.

Related Topics

Computer Science

Biology

Botany

Concepts

Identification (biology) Computer science Biology Botany

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2501.16811
PDF: https://arxiv.org/pdf/2501.16811
OA Status: green
Related Works: 10
OpenAlex ID: https://openalex.org/W4406959683

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4406959683

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2501.16811

Digital Object Identifier
Title: Not Every Patch is Needed: Towards a More Efficient and Effective Backbone for Video-based Person Re-identification

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-01-28

Full publication date if available
Authors: Lanyun Zhu, Tianrun Chen, Deyi Ji, Jieping Ye, Jun Liu

List of authors in order
Landing page: https://arxiv.org/abs/2501.16811

Publisher landing page
PDF URL: https://arxiv.org/pdf/2501.16811

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2501.16811

Direct OA link when available
Concepts: Identification (biology), Computer science, Biology, Botany

Top concepts (fields/topics) attached by OpenAlex
Cited by: 0

Total citation count in OpenAlex
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4406959683
doi	https://doi.org/10.48550/arxiv.2501.16811
ids.doi	https://doi.org/10.48550/arxiv.2501.16811
ids.openalex	https://openalex.org/W4406959683
fwci
type	preprint
title	Not Every Patch is Needed: Towards a More Efficient and Effective Backbone for Video-based Person Re-identification
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10331
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9993000030517578
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1707
topics[0].subfield.display_name	Computer Vision and Pattern Recognition
topics[0].display_name	Video Surveillance and Tracking Methods
topics[1].id	https://openalex.org/T12740
topics[1].field.id	https://openalex.org/fields/22
topics[1].field.display_name	Engineering
topics[1].score	0.9805999994277954
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/2204
topics[1].subfield.display_name	Biomedical Engineering
topics[1].display_name	Gait Recognition and Analysis
topics[2].id	https://openalex.org/T11448
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.9754999876022339
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/1707
topics[2].subfield.display_name	Computer Vision and Pattern Recognition
topics[2].display_name	Face recognition and analysis
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C116834253
concepts[0].level	2
concepts[0].score	0.7687654495239258
concepts[0].wikidata	https://www.wikidata.org/wiki/Q2039217
concepts[0].display_name	Identification (biology)
concepts[1].id	https://openalex.org/C41008148
concepts[1].level	0
concepts[1].score	0.5034505724906921
concepts[1].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[1].display_name	Computer science
concepts[2].id	https://openalex.org/C86803240
concepts[2].level	0
concepts[2].score	0.05026644468307495
concepts[2].wikidata	https://www.wikidata.org/wiki/Q420
concepts[2].display_name	Biology
concepts[3].id	https://openalex.org/C59822182
concepts[3].level	1
concepts[3].score	0.0
concepts[3].wikidata	https://www.wikidata.org/wiki/Q441
concepts[3].display_name	Botany
keywords[0].id	https://openalex.org/keywords/identification
keywords[0].score	0.7687654495239258
keywords[0].display_name	Identification (biology)
keywords[1].id	https://openalex.org/keywords/computer-science
keywords[1].score	0.5034505724906921
keywords[1].display_name	Computer science
keywords[2].id	https://openalex.org/keywords/biology
keywords[2].score	0.05026644468307495
keywords[2].display_name	Biology
language	en
locations[0].id	pmh:oai:arXiv.org:2501.16811
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2501.16811
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2501.16811
locations[1].id	doi:10.48550/arxiv.2501.16811
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2501.16811
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5074956841
authorships[0].author.orcid	https://orcid.org/0000-0001-7309-3330
authorships[0].author.display_name	Lanyun Zhu
authorships[0].author_position	first
authorships[0].raw_author_name	Zhu, Lanyun
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5054714748
authorships[1].author.orcid	https://orcid.org/0000-0003-0177-0157
authorships[1].author.display_name	Tianrun Chen
authorships[1].author_position	middle
authorships[1].raw_author_name	Chen, Tianrun
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5089146499
authorships[2].author.orcid	https://orcid.org/0000-0001-7561-9789
authorships[2].author.display_name	Deyi Ji
authorships[2].author_position	middle
authorships[2].raw_author_name	Ji, Deyi
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5010419481
authorships[3].author.orcid	https://orcid.org/0000-0001-8662-5818
authorships[3].author.display_name	Jieping Ye
authorships[3].author_position	middle
authorships[3].raw_author_name	Ye, Jieping
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5100361668
authorships[4].author.orcid	https://orcid.org/0000-0001-7390-8958
authorships[4].author.display_name	Jun Liu
authorships[4].author_position	last
authorships[4].raw_author_name	Liu, Jun
authorships[4].is_corresponding	False
has_content.pdf	True
has_content.grobid_xml	True
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2501.16811
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	Not Every Patch is Needed: Towards a More Efficient and Effective Backbone for Video-based Person Re-identification
has_fulltext	True
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10331
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9993000030517578
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1707
primary_topic.subfield.display_name	Computer Vision and Pattern Recognition
primary_topic.display_name	Video Surveillance and Tracking Methods
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2390279801, https://openalex.org/W4391913857, https://openalex.org/W2358668433, https://openalex.org/W4396701345, https://openalex.org/W2376932109, https://openalex.org/W2001405890, https://openalex.org/W4396696052
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2501.16811
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2501.16811
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2501.16811
primary_location.id	pmh:oai:arXiv.org:2501.16811
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2501.16811
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2501.16811
publication_date	2025-01-28
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	3, 55, 79, 92, 114
abstract_inverted_index.By	137
abstract_inverted_index.be	46
abstract_inverted_index.by	77, 100, 166
abstract_inverted_index.in	32, 54, 86
abstract_inverted_index.is	84, 179
abstract_inverted_index.of	73, 130
abstract_inverted_index.on	155, 180
abstract_inverted_index.or	22
abstract_inverted_index.to	25, 68, 96, 126, 169, 174
abstract_inverted_index.we	38, 49, 90, 112
abstract_inverted_index.CNN	21
abstract_inverted_index.and	6, 62, 105, 120, 149, 171, 184
abstract_inverted_index.can	144
abstract_inverted_index.due	67
abstract_inverted_index.for	10, 29, 108
abstract_inverted_index.low	150
abstract_inverted_index.new	4, 140
abstract_inverted_index.our	142, 160
abstract_inverted_index.par	181
abstract_inverted_index.the	69, 103, 128, 163, 177
abstract_inverted_index.use	20
abstract_inverted_index.28\%	172
abstract_inverted_index.74\%	167
abstract_inverted_index.ReID	17, 56
abstract_inverted_index.This	0
abstract_inverted_index.both	146
abstract_inverted_index.cost	99, 165
abstract_inverted_index.deep	27
abstract_inverted_index.find	50
abstract_inverted_index.from	134
abstract_inverted_index.high	147
abstract_inverted_index.many	64
abstract_inverted_index.more	80
abstract_inverted_index.only	102
abstract_inverted_index.show	158
abstract_inverted_index.that	40, 51, 118, 159
abstract_inverted_index.this	41, 87
abstract_inverted_index.with	182
abstract_inverted_index.Here,	37
abstract_inverted_index.ViT-B	170, 183
abstract_inverted_index.argue	39
abstract_inverted_index.cost.	152
abstract_inverted_index.could	45
abstract_inverted_index.every	30, 33
abstract_inverted_index.frame	123
abstract_inverted_index.human	74
abstract_inverted_index.issue	129
abstract_inverted_index.novel	115
abstract_inverted_index.often	58
abstract_inverted_index.paper	1
abstract_inverted_index.patch	93
abstract_inverted_index.since	48
abstract_inverted_index.small	60
abstract_inverted_index.these	139
abstract_inverted_index.this,	78
abstract_inverted_index.video	35, 57
abstract_inverted_index.views	132
abstract_inverted_index.while	176
abstract_inverted_index.frame.	36
abstract_inverted_index.frames	53
abstract_inverted_index.global	124
abstract_inverted_index.paper.	88
abstract_inverted_index.person	12
abstract_inverted_index.pseudo	122
abstract_inverted_index.reduce	97
abstract_inverted_index.slight	71
abstract_inverted_index.sparse	135
abstract_inverted_index.(ReID).	14
abstract_inverted_index.achieve	145
abstract_inverted_index.address	127
abstract_inverted_index.beings.	75
abstract_inverted_index.contain	63
abstract_inverted_index.context	125
abstract_inverted_index.crucial	104
abstract_inverted_index.exhibit	59
abstract_inverted_index.extract	26
abstract_inverted_index.feature	43, 109
abstract_inverted_index.inputs.	136
abstract_inverted_index.methods	18
abstract_inverted_index.network	116
abstract_inverted_index.patches	107
abstract_inverted_index.present	113
abstract_inverted_index.reduces	162
abstract_inverted_index.regions	66
abstract_inverted_index.sampled	34
abstract_inverted_index.similar	65
abstract_inverted_index.Inspired	76
abstract_inverted_index.ResNet50	186
abstract_inverted_index.accuracy	178
abstract_inverted_index.approach	161
abstract_inverted_index.backbone	9, 143
abstract_inverted_index.choosing	101
abstract_inverted_index.compared	168, 173
abstract_inverted_index.datasets	157
abstract_inverted_index.designs,	141
abstract_inverted_index.explored	85
abstract_inverted_index.features	28
abstract_inverted_index.multiple	156
abstract_inverted_index.paradigm	83
abstract_inverted_index.position	31
abstract_inverted_index.proposes	2
abstract_inverted_index.utilizes	121
abstract_inverted_index.Extensive	153
abstract_inverted_index.ResNet50,	175
abstract_inverted_index.backbones	24
abstract_inverted_index.different	52
abstract_inverted_index.effective	5
abstract_inverted_index.efficient	7, 82
abstract_inverted_index.generates	119
abstract_inverted_index.introduce	91
abstract_inverted_index.mechanism	95
abstract_inverted_index.movements	72
abstract_inverted_index.resulting	133
abstract_inverted_index.selection	94
abstract_inverted_index.structure	117
abstract_inverted_index.typically	19
abstract_inverted_index.exhaustive	42
abstract_inverted_index.extraction	44
abstract_inverted_index.incomplete	131
abstract_inverted_index.relatively	70
abstract_inverted_index.selective,	81
abstract_inverted_index.differences	61
abstract_inverted_index.experiments	154
abstract_inverted_index.extraction.	110
abstract_inverted_index.outperforms	185
abstract_inverted_index.performance	148
abstract_inverted_index.transformer	23
abstract_inverted_index.video-based	11, 16
abstract_inverted_index.Conventional	15
abstract_inverted_index.unnecessary,	47
abstract_inverted_index.Additionally,	111
abstract_inverted_index.Specifically,	89
abstract_inverted_index.computational	98, 151, 164
abstract_inverted_index.incorporating	138
abstract_inverted_index.plug-and-play	8
abstract_inverted_index.non-repetitive	106
abstract_inverted_index.significantly.	187
abstract_inverted_index.re-identification	13
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	5
citation_normalized_percentile