Personalized Federated Learning for Egocentric Video Gaze Estimation with Comprehensive Parameter Frezzing Article Swipe

PDF

Yuhu Feng , Keisuke Maeda , Takahiro Ogawa , Miki Haseyama ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2502.18123

Egocentric video gaze estimation requires models to capture individual gaze patterns while adapting to diverse user data. Our approach leverages a transformer-based architecture, integrating it into a PFL framework where only the most significant parameters, those exhibiting the highest rate of change during training, are selected and frozen for personalization in client models. Through extensive experimentation on the EGTEA Gaze+ and Ego4D datasets, we demonstrate that FedCPF significantly outperforms previously reported federated learning methods, achieving superior recall, precision, and F1-score. These results confirm the effectiveness of our comprehensive parameters freezing strategy in enhancing model personalization, making FedCPF a promising approach for tasks requiring both adaptability and accuracy in federated learning settings.

Related Topics

Learning Curve

Learning Theory (Education)

Experiential Learning

Practice (Learning Method)

Learning Environment

Machine Learning

Federated States Of Micronesia

Reinforcement Learning

Learning Standards

Attention (Machine Learning)

Learning

Federated State

Concepts

No concepts available.

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2502.18123
PDF: https://arxiv.org/pdf/2502.18123
OA Status: green
OpenAlex ID: https://openalex.org/W4415188183

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4415188183

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2502.18123

Digital Object Identifier
Title: Personalized Federated Learning for Egocentric Video Gaze Estimation with Comprehensive Parameter Frezzing

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-02-25

Full publication date if available
Authors: Yuhu Feng, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama

List of authors in order
Landing page: https://arxiv.org/abs/2502.18123

Publisher landing page
PDF URL: https://arxiv.org/pdf/2502.18123

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/2502.18123

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4415188183
doi	https://doi.org/10.48550/arxiv.2502.18123
ids.doi	https://doi.org/10.48550/arxiv.2502.18123
ids.openalex	https://openalex.org/W4415188183
fwci	0.0
type	preprint
title	Personalized Federated Learning for Egocentric Video Gaze Estimation with Comprehensive Parameter Frezzing
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T13731
topics[0].field.id	https://openalex.org/fields/33
topics[0].field.display_name	Social Sciences
topics[0].score	0.979200005531311
topics[0].domain.id	https://openalex.org/domains/2
topics[0].domain.display_name	Social Sciences
topics[0].subfield.id	https://openalex.org/subfields/3322
topics[0].subfield.display_name	Urban Studies
topics[0].display_name	Advanced Computing and Algorithms
topics[1].id	https://openalex.org/T11707
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.970300018787384
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1709
topics[1].subfield.display_name	Human-Computer Interaction
topics[1].display_name	Gaze Tracking and Assistive Technology
topics[2].id	https://openalex.org/T12702
topics[2].field.id	https://openalex.org/fields/28
topics[2].field.display_name	Neuroscience
topics[2].score	0.9603999853134155
topics[2].domain.id	https://openalex.org/domains/1
topics[2].domain.display_name	Life Sciences
topics[2].subfield.id	https://openalex.org/subfields/2808
topics[2].subfield.display_name	Neurology
topics[2].display_name	Brain Tumor Detection and Classification
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2502.18123
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/2502.18123
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/2502.18123
locations[1].id	doi:10.48550/arxiv.2502.18123
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.2502.18123
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5063121046
authorships[0].author.orcid	https://orcid.org/0009-0006-4819-2066
authorships[0].author.display_name	Yuhu Feng
authorships[0].author_position	first
authorships[0].raw_author_name	Feng, Yuhu
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5033215072
authorships[1].author.orcid	https://orcid.org/0000-0001-8039-3462
authorships[1].author.display_name	Keisuke Maeda
authorships[1].author_position	middle
authorships[1].raw_author_name	Maeda, Keisuke
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5009032240
authorships[2].author.orcid	https://orcid.org/0000-0001-5332-8112
authorships[2].author.display_name	Takahiro Ogawa
authorships[2].author_position	middle
authorships[2].raw_author_name	Ogawa, Takahiro
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5063903016
authorships[3].author.orcid	https://orcid.org/0000-0003-1496-1761
authorships[3].author.display_name	Miki Haseyama
authorships[3].author_position	last
authorships[3].raw_author_name	Haseyama, Miki
authorships[3].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2502.18123
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-15T00:00:00
display_name	Personalized Federated Learning for Egocentric Video Gaze Estimation with Comprehensive Parameter Frezzing
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T13731
primary_topic.field.id	https://openalex.org/fields/33
primary_topic.field.display_name	Social Sciences
primary_topic.score	0.979200005531311
primary_topic.domain.id	https://openalex.org/domains/2
primary_topic.domain.display_name	Social Sciences
primary_topic.subfield.id	https://openalex.org/subfields/3322
primary_topic.subfield.display_name	Urban Studies
primary_topic.display_name	Advanced Computing and Algorithms
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2502.18123
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/2502.18123
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/2502.18123
primary_location.id	pmh:oai:arXiv.org:2502.18123
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/2502.18123
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/2502.18123
publication_date	2025-02-25
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	20, 26, 97
abstract_inverted_index.in	50, 91, 107
abstract_inverted_index.it	24
abstract_inverted_index.of	40, 85
abstract_inverted_index.on	56
abstract_inverted_index.to	6, 13
abstract_inverted_index.we	63
abstract_inverted_index.Our	17
abstract_inverted_index.PFL	27
abstract_inverted_index.and	46, 60, 78, 105
abstract_inverted_index.are	44
abstract_inverted_index.for	48, 100
abstract_inverted_index.our	86
abstract_inverted_index.the	31, 37, 57, 83
abstract_inverted_index.both	103
abstract_inverted_index.gaze	2, 9
abstract_inverted_index.into	25
abstract_inverted_index.most	32
abstract_inverted_index.only	30
abstract_inverted_index.rate	39
abstract_inverted_index.that	65
abstract_inverted_index.user	15
abstract_inverted_index.EGTEA	58
abstract_inverted_index.Ego4D	61
abstract_inverted_index.Gaze+	59
abstract_inverted_index.These	80
abstract_inverted_index.data.	16
abstract_inverted_index.model	93
abstract_inverted_index.tasks	101
abstract_inverted_index.those	35
abstract_inverted_index.video	1
abstract_inverted_index.where	29
abstract_inverted_index.while	11
abstract_inverted_index.FedCPF	66, 96
abstract_inverted_index.change	41
abstract_inverted_index.client	51
abstract_inverted_index.during	42
abstract_inverted_index.frozen	47
abstract_inverted_index.making	95
abstract_inverted_index.models	5
abstract_inverted_index.Through	53
abstract_inverted_index.capture	7
abstract_inverted_index.confirm	82
abstract_inverted_index.diverse	14
abstract_inverted_index.highest	38
abstract_inverted_index.models.	52
abstract_inverted_index.recall,	76
abstract_inverted_index.results	81
abstract_inverted_index.accuracy	106
abstract_inverted_index.adapting	12
abstract_inverted_index.approach	18, 99
abstract_inverted_index.freezing	89
abstract_inverted_index.learning	72, 109
abstract_inverted_index.methods,	73
abstract_inverted_index.patterns	10
abstract_inverted_index.reported	70
abstract_inverted_index.requires	4
abstract_inverted_index.selected	45
abstract_inverted_index.strategy	90
abstract_inverted_index.superior	75
abstract_inverted_index.F1-score.	79
abstract_inverted_index.achieving	74
abstract_inverted_index.datasets,	62
abstract_inverted_index.enhancing	92
abstract_inverted_index.extensive	54
abstract_inverted_index.federated	71, 108
abstract_inverted_index.framework	28
abstract_inverted_index.leverages	19
abstract_inverted_index.promising	98
abstract_inverted_index.requiring	102
abstract_inverted_index.settings.	110
abstract_inverted_index.training,	43
abstract_inverted_index.Egocentric	0
abstract_inverted_index.estimation	3
abstract_inverted_index.exhibiting	36
abstract_inverted_index.individual	8
abstract_inverted_index.parameters	88
abstract_inverted_index.precision,	77
abstract_inverted_index.previously	69
abstract_inverted_index.demonstrate	64
abstract_inverted_index.integrating	23
abstract_inverted_index.outperforms	68
abstract_inverted_index.parameters,	34
abstract_inverted_index.significant	33
abstract_inverted_index.adaptability	104
abstract_inverted_index.architecture,	22
abstract_inverted_index.comprehensive	87
abstract_inverted_index.effectiveness	84
abstract_inverted_index.significantly	67
abstract_inverted_index.experimentation	55
abstract_inverted_index.personalization	49
abstract_inverted_index.personalization,	94
abstract_inverted_index.transformer-based	21
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	4
citation_normalized_percentile