TwinLiteNet+: An Enhanced Multi-Task Segmentation Model for Autonomous Driving Article Swipe

PDF

Quang-Huy Che , Duc-Tri Le , Minh-Quan Pham , Vinh-Tiep Nguyen , D. Lam ·

YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2403.16958

Semantic segmentation is a fundamental perception task in autonomous driving, particularly for identifying drivable areas and lane markings to enable safe navigation. However, most state-of-the-art (SOTA) models are computationally intensive and unsuitable for real-time deployment on resource-constrained embedded devices. In this paper, we introduce TwinLiteNet+, an enhanced multi-task segmentation model designed for real-time drivable area and lane segmentation with high efficiency. TwinLiteNet+ employs a hybrid encoder architecture that integrates stride-based dilated convolutions and depthwise separable dilated convolutions, balancing representational capacity and computational cost. To improve task-specific decoding, we propose two lightweight upsampling modules-Upper Convolution Block (UCB) and Upper Simple Block (USB)-alongside a Partial Class Activation Attention (PCAA) mechanism that enhances segmentation precision. The model is available in four configurations, ranging from the ultra-compact TwinLiteNet+_{Nano} (34K parameters) to the high-performance TwinLiteNet+_{Large} (1.94M parameters). On the BDD100K dataset, TwinLiteNet+_{Large} achieves 92.9% mIoU for drivable area segmentation and 34.2% IoU for lane segmentation-surpassing existing state-of-the-art models while requiring 11x fewer floating-point operations (FLOPs) for computation. Extensive evaluations on embedded devices demonstrate superior inference speed, quantization robustness (INT8/FP16), and energy efficiency, validating TwinLiteNet+ as a compelling solution for real-world autonomous driving systems. Code is available at https://github.com/chequanghuy/TwinLiteNetPlus.

Related Topics

Segmentation Fault

Computer Vision

Computer Science

Artificial Intelligence

Concepts

Segmentation Computer vision Computer science Artificial intelligence

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2403.16958
PDF: https://arxiv.org/pdf/2403.16958
OA Status: green
Related Works: 10
OpenAlex ID: https://openalex.org/W4393213804

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4393213804

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2403.16958

Digital Object Identifier
Title: TwinLiteNet+: An Enhanced Multi-Task Segmentation Model for Autonomous Driving

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2024

Year of publication
Publication date: 2024-03-25

Full publication date if available
Authors: Quang-Huy Che, Duc-Tri Le, Minh-Quan Pham, Vinh-Tiep Nguyen, D. Lam

List of authors in order
Landing page: https://arxiv.org/abs/2403.16958

Publisher landing page
PDF URL: https://arxiv.org/pdf/2403.16958

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/2403.16958

Direct OA link when available
Concepts: Segmentation, Computer vision, Computer science, Artificial intelligence

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/W4393213804
doi	https://doi.org/10.48550/arxiv.2403.16958
ids.doi	https://doi.org/10.48550/arxiv.2403.16958
ids.openalex	https://openalex.org/W4393213804
fwci
type	preprint
title	TwinLiteNet+: An Enhanced Multi-Task Segmentation Model for Autonomous Driving
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T11099
topics[0].field.id	https://openalex.org/fields/22
topics[0].field.display_name	Engineering
topics[0].score	0.9990000128746033
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/2203
topics[0].subfield.display_name	Automotive Engineering
topics[0].display_name	Autonomous Vehicle Technology and Safety
topics[1].id	https://openalex.org/T10331
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.996999979019165
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1707
topics[1].subfield.display_name	Computer Vision and Pattern Recognition
topics[1].display_name	Video Surveillance and Tracking Methods
topics[2].id	https://openalex.org/T13282
topics[2].field.id	https://openalex.org/fields/22
topics[2].field.display_name	Engineering
topics[2].score	0.993399977684021
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/2212
topics[2].subfield.display_name	Ocean Engineering
topics[2].display_name	Automated Road and Building Extraction
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C89600930
concepts[0].level	2
concepts[0].score	0.7328256368637085
concepts[0].wikidata	https://www.wikidata.org/wiki/Q1423946
concepts[0].display_name	Segmentation
concepts[1].id	https://openalex.org/C31972630
concepts[1].level	1
concepts[1].score	0.5222055315971375
concepts[1].wikidata	https://www.wikidata.org/wiki/Q844240
concepts[1].display_name	Computer vision
concepts[2].id	https://openalex.org/C41008148
concepts[2].level	0
concepts[2].score	0.498624324798584
concepts[2].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[2].display_name	Computer science
concepts[3].id	https://openalex.org/C154945302
concepts[3].level	1
concepts[3].score	0.4793427884578705
concepts[3].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[3].display_name	Artificial intelligence
keywords[0].id	https://openalex.org/keywords/segmentation
keywords[0].score	0.7328256368637085
keywords[0].display_name	Segmentation
keywords[1].id	https://openalex.org/keywords/computer-vision
keywords[1].score	0.5222055315971375
keywords[1].display_name	Computer vision
keywords[2].id	https://openalex.org/keywords/computer-science
keywords[2].score	0.498624324798584
keywords[2].display_name	Computer science
keywords[3].id	https://openalex.org/keywords/artificial-intelligence
keywords[3].score	0.4793427884578705
keywords[3].display_name	Artificial intelligence
language	en
locations[0].id	pmh:oai:arXiv.org:2403.16958
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/2403.16958
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/2403.16958
locations[1].id	doi:10.48550/arxiv.2403.16958
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.2403.16958
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5094236163
authorships[0].author.orcid
authorships[0].author.display_name	Quang-Huy Che
authorships[0].author_position	first
authorships[0].raw_author_name	Che, Quang-Huy
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5101323915
authorships[1].author.orcid
authorships[1].author.display_name	Duc-Tri Le
authorships[1].author_position	middle
authorships[1].raw_author_name	Le, Duc-Tri
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5102771277
authorships[2].author.orcid	https://orcid.org/0000-0002-4389-6869
authorships[2].author.display_name	Minh-Quan Pham
authorships[2].author_position	middle
authorships[2].raw_author_name	Pham, Minh-Quan
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5054588832
authorships[3].author.orcid	https://orcid.org/0000-0003-4260-7874
authorships[3].author.display_name	Vinh-Tiep Nguyen
authorships[3].author_position	middle
authorships[3].raw_author_name	Nguyen, Vinh-Tiep
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5113153300
authorships[4].author.orcid
authorships[4].author.display_name	D. Lam
authorships[4].author_position	last
authorships[4].raw_author_name	Lam, Duc-Khai
authorships[4].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/2403.16958
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2024-03-27T00:00:00
display_name	TwinLiteNet+: An Enhanced Multi-Task Segmentation Model for Autonomous Driving
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T11099
primary_topic.field.id	https://openalex.org/fields/22
primary_topic.field.display_name	Engineering
primary_topic.score	0.9990000128746033
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/2203
primary_topic.subfield.display_name	Automotive Engineering
primary_topic.display_name	Autonomous Vehicle Technology and Safety
related_works	https://openalex.org/W2058170566, https://openalex.org/W2755342338, https://openalex.org/W2772917594, https://openalex.org/W2775347418, https://openalex.org/W2166024367, https://openalex.org/W3116076068, https://openalex.org/W2229312674, https://openalex.org/W2951359407, https://openalex.org/W2079911747, https://openalex.org/W1969923398
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2403.16958
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/2403.16958
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/2403.16958
primary_location.id	pmh:oai:arXiv.org:2403.16958
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/2403.16958
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/2403.16958
publication_date	2024-03-25
publication_year	2024
referenced_works_count	0
abstract_inverted_index.a	3, 63, 101, 180
abstract_inverted_index.In	39
abstract_inverted_index.On	132
abstract_inverted_index.To	83
abstract_inverted_index.an	45
abstract_inverted_index.as	179
abstract_inverted_index.at	191
abstract_inverted_index.in	7, 116
abstract_inverted_index.is	2, 114, 189
abstract_inverted_index.on	35, 164
abstract_inverted_index.to	18, 126
abstract_inverted_index.we	42, 87
abstract_inverted_index.11x	155
abstract_inverted_index.IoU	146
abstract_inverted_index.The	112
abstract_inverted_index.and	15, 30, 55, 72, 80, 96, 144, 174
abstract_inverted_index.are	27
abstract_inverted_index.for	11, 32, 51, 140, 147, 160, 183
abstract_inverted_index.the	121, 127, 133
abstract_inverted_index.two	89
abstract_inverted_index.(34K	124
abstract_inverted_index.Code	188
abstract_inverted_index.area	54, 142
abstract_inverted_index.four	117
abstract_inverted_index.from	120
abstract_inverted_index.high	59
abstract_inverted_index.lane	16, 56, 148
abstract_inverted_index.mIoU	139
abstract_inverted_index.most	23
abstract_inverted_index.safe	20
abstract_inverted_index.task	6
abstract_inverted_index.that	67, 108
abstract_inverted_index.this	40
abstract_inverted_index.with	58
abstract_inverted_index.(UCB)	95
abstract_inverted_index.34.2%	145
abstract_inverted_index.92.9%	138
abstract_inverted_index.Block	94, 99
abstract_inverted_index.Class	103
abstract_inverted_index.Upper	97
abstract_inverted_index.areas	14
abstract_inverted_index.cost.	82
abstract_inverted_index.fewer	156
abstract_inverted_index.model	49, 113
abstract_inverted_index.while	153
abstract_inverted_index.(1.94M	130
abstract_inverted_index.(PCAA)	106
abstract_inverted_index.(SOTA)	25
abstract_inverted_index.Simple	98
abstract_inverted_index.enable	19
abstract_inverted_index.energy	175
abstract_inverted_index.hybrid	64
abstract_inverted_index.models	26, 152
abstract_inverted_index.paper,	41
abstract_inverted_index.speed,	170
abstract_inverted_index.(FLOPs)	159
abstract_inverted_index.BDD100K	134
abstract_inverted_index.Partial	102
abstract_inverted_index.devices	166
abstract_inverted_index.dilated	70, 75
abstract_inverted_index.driving	186
abstract_inverted_index.employs	62
abstract_inverted_index.encoder	65
abstract_inverted_index.improve	84
abstract_inverted_index.propose	88
abstract_inverted_index.ranging	119
abstract_inverted_index.However,	22
abstract_inverted_index.Semantic	0
abstract_inverted_index.achieves	137
abstract_inverted_index.capacity	79
abstract_inverted_index.dataset,	135
abstract_inverted_index.designed	50
abstract_inverted_index.devices.	38
abstract_inverted_index.drivable	13, 53, 141
abstract_inverted_index.driving,	9
abstract_inverted_index.embedded	37, 165
abstract_inverted_index.enhanced	46
abstract_inverted_index.enhances	109
abstract_inverted_index.existing	150
abstract_inverted_index.markings	17
abstract_inverted_index.solution	182
abstract_inverted_index.superior	168
abstract_inverted_index.systems.	187
abstract_inverted_index.Attention	105
abstract_inverted_index.Extensive	162
abstract_inverted_index.available	115, 190
abstract_inverted_index.balancing	77
abstract_inverted_index.decoding,	86
abstract_inverted_index.depthwise	73
abstract_inverted_index.inference	169
abstract_inverted_index.intensive	29
abstract_inverted_index.introduce	43
abstract_inverted_index.mechanism	107
abstract_inverted_index.real-time	33, 52
abstract_inverted_index.requiring	154
abstract_inverted_index.separable	74
abstract_inverted_index.Activation	104
abstract_inverted_index.autonomous	8, 185
abstract_inverted_index.compelling	181
abstract_inverted_index.deployment	34
abstract_inverted_index.integrates	68
abstract_inverted_index.multi-task	47
abstract_inverted_index.operations	158
abstract_inverted_index.perception	5
abstract_inverted_index.precision.	111
abstract_inverted_index.real-world	184
abstract_inverted_index.robustness	172
abstract_inverted_index.unsuitable	31
abstract_inverted_index.upsampling	91
abstract_inverted_index.validating	177
abstract_inverted_index.Convolution	93
abstract_inverted_index.demonstrate	167
abstract_inverted_index.efficiency,	176
abstract_inverted_index.efficiency.	60
abstract_inverted_index.evaluations	163
abstract_inverted_index.fundamental	4
abstract_inverted_index.identifying	12
abstract_inverted_index.lightweight	90
abstract_inverted_index.navigation.	21
abstract_inverted_index.parameters)	125
abstract_inverted_index.(INT8/FP16),	173
abstract_inverted_index.TwinLiteNet+	61, 178
abstract_inverted_index.architecture	66
abstract_inverted_index.computation.	161
abstract_inverted_index.convolutions	71
abstract_inverted_index.parameters).	131
abstract_inverted_index.particularly	10
abstract_inverted_index.quantization	171
abstract_inverted_index.segmentation	1, 48, 57, 110, 143
abstract_inverted_index.stride-based	69
abstract_inverted_index.TwinLiteNet+,	44
abstract_inverted_index.computational	81
abstract_inverted_index.convolutions,	76
abstract_inverted_index.modules-Upper	92
abstract_inverted_index.task-specific	85
abstract_inverted_index.ultra-compact	122
abstract_inverted_index.floating-point	157
abstract_inverted_index.(USB)-alongside	100
abstract_inverted_index.computationally	28
abstract_inverted_index.configurations,	118
abstract_inverted_index.high-performance	128
abstract_inverted_index.representational	78
abstract_inverted_index.state-of-the-art	24, 151
abstract_inverted_index.TwinLiteNet+_{Nano}	123
abstract_inverted_index.TwinLiteNet+_{Large}	129, 136
abstract_inverted_index.resource-constrained	36
abstract_inverted_index.segmentation-surpassing	149
abstract_inverted_index.https://github.com/chequanghuy/TwinLiteNetPlus.	192
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	5
citation_normalized_percentile