Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications Article Swipe

PDF

Yuwen Xiong , Zhiqi Li , Yuntao Chen , Feng Wang , Xizhou Zhu , Jiapeng Luo , Wenhai Wang , Tong Lü , Hongsheng Li , Yu Qiao , Lewei Lu , Jie Zhou , Jifeng Dai ·

YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2401.06197

We introduce Deformable Convolution v4 (DCNv4), a highly efficient and effective operator designed for a broad spectrum of vision applications. DCNv4 addresses the limitations of its predecessor, DCNv3, with two key enhancements: 1. removing softmax normalization in spatial aggregation to enhance its dynamic property and expressive power and 2. optimizing memory access to minimize redundant operations for speedup. These improvements result in a significantly faster convergence compared to DCNv3 and a substantial increase in processing speed, with DCNv4 achieving more than three times the forward speed. DCNv4 demonstrates exceptional performance across various tasks, including image classification, instance and semantic segmentation, and notably, image generation. When integrated into generative models like U-Net in the latent diffusion model, DCNv4 outperforms its baseline, underscoring its possibility to enhance generative models. In practical applications, replacing DCNv3 with DCNv4 in the InternImage model to create FlashInternImage results in up to 80% speed increase and further performance improvement without further modifications. The advancements in speed and efficiency of DCNv4, combined with its robust performance across diverse vision tasks, show its potential as a foundational building block for future vision models.

Related Topics

Computer Science

Softmax Function

Artificial Intelligence

Operations Management

Concepts

Speedup Computer science Softmax function Artificial intelligence Block (permutation group theory) Operator (biology) Normalization (sociology) Performance improvement Generative model Machine learning Computer vision Generative grammar Deep learning Parallel computing Operations management Repressor Anthropology Chemistry Geometry Mathematics Biochemistry Transcription factor Economics Gene Sociology

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2401.06197
PDF: https://arxiv.org/pdf/2401.06197
OA Status: green
Cited By: 8
Related Works: 10
OpenAlex ID: https://openalex.org/W4390897321

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4390897321

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2401.06197

Digital Object Identifier
Title: Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2024

Year of publication
Publication date: 2024-01-11

Full publication date if available
Authors: Yuwen Xiong, Zhiqi Li, Yuntao Chen, Feng Wang, Xizhou Zhu, Jiapeng Luo, Wenhai Wang, Tong Lü, Hongsheng Li, Yu Qiao, Lewei Lu, Jie Zhou, Jifeng Dai

List of authors in order
Landing page: https://arxiv.org/abs/2401.06197

Publisher landing page
PDF URL: https://arxiv.org/pdf/2401.06197

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/2401.06197

Direct OA link when available
Concepts: Speedup, Computer science, Softmax function, Artificial intelligence, Block (permutation group theory), Operator (biology), Normalization (sociology), Performance improvement, Generative model, Machine learning, Computer vision, Generative grammar, Deep learning, Parallel computing, Operations management, Repressor, Anthropology, Chemistry, Geometry, Mathematics, Biochemistry, Transcription factor, Economics, Gene, Sociology

Top concepts (fields/topics) attached by OpenAlex
Cited by: 8

Total citation count in OpenAlex
Citations by year (recent): 2025: 7, 2024: 1

Per-year citation counts (last 5 years)
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4390897321
doi	https://doi.org/10.48550/arxiv.2401.06197
ids.doi	https://doi.org/10.48550/arxiv.2401.06197
ids.openalex	https://openalex.org/W4390897321
fwci
type	preprint
title	Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10036
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.9994000196456909
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	Advanced Neural Network Applications
topics[1].id	https://openalex.org/T10627
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.9988999962806702
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	Advanced Image and Video Retrieval Techniques
topics[2].id	https://openalex.org/T11307
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.9943000078201294
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/1702
topics[2].subfield.display_name	Artificial Intelligence
topics[2].display_name	Domain Adaptation and Few-Shot Learning
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C68339613
concepts[0].level	2
concepts[0].score	0.8497605323791504
concepts[0].wikidata	https://www.wikidata.org/wiki/Q1549489
concepts[0].display_name	Speedup
concepts[1].id	https://openalex.org/C41008148
concepts[1].level	0
concepts[1].score	0.805129885673523
concepts[1].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[1].display_name	Computer science
concepts[2].id	https://openalex.org/C188441871
concepts[2].level	3
concepts[2].score	0.7577035427093506
concepts[2].wikidata	https://www.wikidata.org/wiki/Q7554146
concepts[2].display_name	Softmax function
concepts[3].id	https://openalex.org/C154945302
concepts[3].level	1
concepts[3].score	0.497997522354126
concepts[3].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[3].display_name	Artificial intelligence
concepts[4].id	https://openalex.org/C2777210771
concepts[4].level	2
concepts[4].score	0.47114983201026917
concepts[4].wikidata	https://www.wikidata.org/wiki/Q4927124
concepts[4].display_name	Block (permutation group theory)
concepts[5].id	https://openalex.org/C17020691
concepts[5].level	5
concepts[5].score	0.43885916471481323
concepts[5].wikidata	https://www.wikidata.org/wiki/Q139677
concepts[5].display_name	Operator (biology)
concepts[6].id	https://openalex.org/C136886441
concepts[6].level	2
concepts[6].score	0.42511457204818726
concepts[6].wikidata	https://www.wikidata.org/wiki/Q926129
concepts[6].display_name	Normalization (sociology)
concepts[7].id	https://openalex.org/C2778915421
concepts[7].level	2
concepts[7].score	0.4238213002681732
concepts[7].wikidata	https://www.wikidata.org/wiki/Q3643177
concepts[7].display_name	Performance improvement
concepts[8].id	https://openalex.org/C167966045
concepts[8].level	3
concepts[8].score	0.41874960064888
concepts[8].wikidata	https://www.wikidata.org/wiki/Q5532625
concepts[8].display_name	Generative model
concepts[9].id	https://openalex.org/C119857082
concepts[9].level	1
concepts[9].score	0.37738850712776184
concepts[9].wikidata	https://www.wikidata.org/wiki/Q2539
concepts[9].display_name	Machine learning
concepts[10].id	https://openalex.org/C31972630
concepts[10].level	1
concepts[10].score	0.32760319113731384
concepts[10].wikidata	https://www.wikidata.org/wiki/Q844240
concepts[10].display_name	Computer vision
concepts[11].id	https://openalex.org/C39890363
concepts[11].level	2
concepts[11].score	0.29964447021484375
concepts[11].wikidata	https://www.wikidata.org/wiki/Q36108
concepts[11].display_name	Generative grammar
concepts[12].id	https://openalex.org/C108583219
concepts[12].level	2
concepts[12].score	0.24680376052856445
concepts[12].wikidata	https://www.wikidata.org/wiki/Q197536
concepts[12].display_name	Deep learning
concepts[13].id	https://openalex.org/C173608175
concepts[13].level	1
concepts[13].score	0.14250493049621582
concepts[13].wikidata	https://www.wikidata.org/wiki/Q232661
concepts[13].display_name	Parallel computing
concepts[14].id	https://openalex.org/C21547014
concepts[14].level	1
concepts[14].score	0.0
concepts[14].wikidata	https://www.wikidata.org/wiki/Q1423657
concepts[14].display_name	Operations management
concepts[15].id	https://openalex.org/C158448853
concepts[15].level	4
concepts[15].score	0.0
concepts[15].wikidata	https://www.wikidata.org/wiki/Q425218
concepts[15].display_name	Repressor
concepts[16].id	https://openalex.org/C19165224
concepts[16].level	1
concepts[16].score	0.0
concepts[16].wikidata	https://www.wikidata.org/wiki/Q23404
concepts[16].display_name	Anthropology
concepts[17].id	https://openalex.org/C185592680
concepts[17].level	0
concepts[17].score	0.0
concepts[17].wikidata	https://www.wikidata.org/wiki/Q2329
concepts[17].display_name	Chemistry
concepts[18].id	https://openalex.org/C2524010
concepts[18].level	1
concepts[18].score	0.0
concepts[18].wikidata	https://www.wikidata.org/wiki/Q8087
concepts[18].display_name	Geometry
concepts[19].id	https://openalex.org/C33923547
concepts[19].level	0
concepts[19].score	0.0
concepts[19].wikidata	https://www.wikidata.org/wiki/Q395
concepts[19].display_name	Mathematics
concepts[20].id	https://openalex.org/C55493867
concepts[20].level	1
concepts[20].score	0.0
concepts[20].wikidata	https://www.wikidata.org/wiki/Q7094
concepts[20].display_name	Biochemistry
concepts[21].id	https://openalex.org/C86339819
concepts[21].level	3
concepts[21].score	0.0
concepts[21].wikidata	https://www.wikidata.org/wiki/Q407384
concepts[21].display_name	Transcription factor
concepts[22].id	https://openalex.org/C162324750
concepts[22].level	0
concepts[22].score	0.0
concepts[22].wikidata	https://www.wikidata.org/wiki/Q8134
concepts[22].display_name	Economics
concepts[23].id	https://openalex.org/C104317684
concepts[23].level	2
concepts[23].score	0.0
concepts[23].wikidata	https://www.wikidata.org/wiki/Q7187
concepts[23].display_name	Gene
concepts[24].id	https://openalex.org/C144024400
concepts[24].level	0
concepts[24].score	0.0
concepts[24].wikidata	https://www.wikidata.org/wiki/Q21201
concepts[24].display_name	Sociology
keywords[0].id	https://openalex.org/keywords/speedup
keywords[0].score	0.8497605323791504
keywords[0].display_name	Speedup
keywords[1].id	https://openalex.org/keywords/computer-science
keywords[1].score	0.805129885673523
keywords[1].display_name	Computer science
keywords[2].id	https://openalex.org/keywords/softmax-function
keywords[2].score	0.7577035427093506
keywords[2].display_name	Softmax function
keywords[3].id	https://openalex.org/keywords/artificial-intelligence
keywords[3].score	0.497997522354126
keywords[3].display_name	Artificial intelligence
keywords[4].id	https://openalex.org/keywords/block
keywords[4].score	0.47114983201026917
keywords[4].display_name	Block (permutation group theory)
keywords[5].id	https://openalex.org/keywords/operator
keywords[5].score	0.43885916471481323
keywords[5].display_name	Operator (biology)
keywords[6].id	https://openalex.org/keywords/normalization
keywords[6].score	0.42511457204818726
keywords[6].display_name	Normalization (sociology)
keywords[7].id	https://openalex.org/keywords/performance-improvement
keywords[7].score	0.4238213002681732
keywords[7].display_name	Performance improvement
keywords[8].id	https://openalex.org/keywords/generative-model
keywords[8].score	0.41874960064888
keywords[8].display_name	Generative model
keywords[9].id	https://openalex.org/keywords/machine-learning
keywords[9].score	0.37738850712776184
keywords[9].display_name	Machine learning
keywords[10].id	https://openalex.org/keywords/computer-vision
keywords[10].score	0.32760319113731384
keywords[10].display_name	Computer vision
keywords[11].id	https://openalex.org/keywords/generative-grammar
keywords[11].score	0.29964447021484375
keywords[11].display_name	Generative grammar
keywords[12].id	https://openalex.org/keywords/deep-learning
keywords[12].score	0.24680376052856445
keywords[12].display_name	Deep learning
keywords[13].id	https://openalex.org/keywords/parallel-computing
keywords[13].score	0.14250493049621582
keywords[13].display_name	Parallel computing
language	en
locations[0].id	pmh:oai:arXiv.org:2401.06197
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/2401.06197
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/2401.06197
locations[1].id	doi:10.48550/arxiv.2401.06197
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.2401.06197
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5100528261
authorships[0].author.orcid
authorships[0].author.display_name	Yuwen Xiong
authorships[0].author_position	first
authorships[0].raw_author_name	Xiong, Yuwen
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5100675974
authorships[1].author.orcid	https://orcid.org/0000-0002-7041-0748
authorships[1].author.display_name	Zhiqi Li
authorships[1].author_position	middle
authorships[1].raw_author_name	Li, Zhiqi
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5102821179
authorships[2].author.orcid	https://orcid.org/0000-0002-9555-1897
authorships[2].author.display_name	Yuntao Chen
authorships[2].author_position	middle
authorships[2].raw_author_name	Chen, Yuntao
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5100431109
authorships[3].author.orcid	https://orcid.org/0000-0001-5457-4497
authorships[3].author.display_name	Feng Wang
authorships[3].author_position	middle
authorships[3].raw_author_name	Wang, Feng
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5112877229
authorships[4].author.orcid
authorships[4].author.display_name	Xizhou Zhu
authorships[4].author_position	middle
authorships[4].raw_author_name	Zhu, Xizhou
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5079292022
authorships[5].author.orcid	https://orcid.org/0000-0002-5585-8632
authorships[5].author.display_name	Jiapeng Luo
authorships[5].author_position	middle
authorships[5].raw_author_name	Luo, Jiapeng
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5101827340
authorships[6].author.orcid	https://orcid.org/0000-0002-2418-3134
authorships[6].author.display_name	Wenhai Wang
authorships[6].author_position	middle
authorships[6].raw_author_name	Wang, Wenhai
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5061696740
authorships[7].author.orcid	https://orcid.org/0000-0002-7051-5347
authorships[7].author.display_name	Tong Lü
authorships[7].author_position	middle
authorships[7].raw_author_name	Lu, Tong
authorships[7].is_corresponding	False
authorships[8].author.id	https://openalex.org/A5100732450
authorships[8].author.orcid	https://orcid.org/0000-0002-2664-7975
authorships[8].author.display_name	Hongsheng Li
authorships[8].author_position	middle
authorships[8].raw_author_name	Li, Hongsheng
authorships[8].is_corresponding	False
authorships[9].author.id	https://openalex.org/A5100748135
authorships[9].author.orcid	https://orcid.org/0000-0002-1889-2567
authorships[9].author.display_name	Yu Qiao
authorships[9].author_position	middle
authorships[9].raw_author_name	Qiao, Yu
authorships[9].is_corresponding	False
authorships[10].author.id	https://openalex.org/A5000474748
authorships[10].author.orcid	https://orcid.org/0009-0009-9809-3818
authorships[10].author.display_name	Lewei Lu
authorships[10].author_position	middle
authorships[10].raw_author_name	Lu, Lewei
authorships[10].is_corresponding	False
authorships[11].author.id	https://openalex.org/A5100620306
authorships[11].author.orcid	https://orcid.org/0000-0001-7701-234X
authorships[11].author.display_name	Jie Zhou
authorships[11].author_position	middle
authorships[11].raw_author_name	Zhou, Jie
authorships[11].is_corresponding	False
authorships[12].author.id	https://openalex.org/A5026944066
authorships[12].author.orcid	https://orcid.org/0000-0002-6785-0785
authorships[12].author.display_name	Jifeng Dai
authorships[12].author_position	last
authorships[12].raw_author_name	Dai, Jifeng
authorships[12].is_corresponding	False
has_content.pdf	True
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2401.06197
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2024-01-16T00:00:00
display_name	Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10036
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.9994000196456909
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	Advanced Neural Network Applications
related_works	https://openalex.org/W3107204728, https://openalex.org/W4287591324, https://openalex.org/W2980176872, https://openalex.org/W4226420367, https://openalex.org/W2962876041, https://openalex.org/W3090555870, https://openalex.org/W3108503355, https://openalex.org/W2249953602, https://openalex.org/W2912971006, https://openalex.org/W4384937382
cited_by_count	8
counts_by_year[0].year	2025
counts_by_year[0].cited_by_count	7
counts_by_year[1].year	2024
counts_by_year[1].cited_by_count	1
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2401.06197
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/2401.06197
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/2401.06197
primary_location.id	pmh:oai:arXiv.org:2401.06197
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/2401.06197
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/2401.06197
publication_date	2024-01-11
publication_year	2024
referenced_works_count	0
abstract_inverted_index.a	6, 14, 62, 70, 176
abstract_inverted_index.1.	32
abstract_inverted_index.2.	48
abstract_inverted_index.In	127
abstract_inverted_index.We	0
abstract_inverted_index.as	175
abstract_inverted_index.in	36, 61, 73, 111, 134, 142, 157
abstract_inverted_index.of	17, 24, 161
abstract_inverted_index.to	39, 52, 67, 123, 138, 144
abstract_inverted_index.up	143
abstract_inverted_index.v4	4
abstract_inverted_index.80%	145
abstract_inverted_index.The	155
abstract_inverted_index.and	9, 44, 47, 69, 97, 100, 148, 159
abstract_inverted_index.for	13, 56, 180
abstract_inverted_index.its	25, 41, 118, 121, 165, 173
abstract_inverted_index.key	30
abstract_inverted_index.the	22, 83, 112, 135
abstract_inverted_index.two	29
abstract_inverted_index.When	104
abstract_inverted_index.into	106
abstract_inverted_index.like	109
abstract_inverted_index.more	79
abstract_inverted_index.show	172
abstract_inverted_index.than	80
abstract_inverted_index.with	28, 76, 132, 164
abstract_inverted_index.DCNv3	68, 131
abstract_inverted_index.DCNv4	20, 77, 86, 116, 133
abstract_inverted_index.These	58
abstract_inverted_index.U-Net	110
abstract_inverted_index.block	179
abstract_inverted_index.broad	15
abstract_inverted_index.image	94, 102
abstract_inverted_index.model	137
abstract_inverted_index.power	46
abstract_inverted_index.speed	146, 158
abstract_inverted_index.three	81
abstract_inverted_index.times	82
abstract_inverted_index.DCNv3,	27
abstract_inverted_index.DCNv4,	162
abstract_inverted_index.access	51
abstract_inverted_index.across	90, 168
abstract_inverted_index.create	139
abstract_inverted_index.faster	64
abstract_inverted_index.future	181
abstract_inverted_index.highly	7
abstract_inverted_index.latent	113
abstract_inverted_index.memory	50
abstract_inverted_index.model,	115
abstract_inverted_index.models	108
abstract_inverted_index.result	60
abstract_inverted_index.robust	166
abstract_inverted_index.speed,	75
abstract_inverted_index.speed.	85
abstract_inverted_index.tasks,	92, 171
abstract_inverted_index.vision	18, 170, 182
abstract_inverted_index.diverse	169
abstract_inverted_index.dynamic	42
abstract_inverted_index.enhance	40, 124
abstract_inverted_index.forward	84
abstract_inverted_index.further	149, 153
abstract_inverted_index.models.	126, 183
abstract_inverted_index.results	141
abstract_inverted_index.softmax	34
abstract_inverted_index.spatial	37
abstract_inverted_index.various	91
abstract_inverted_index.without	152
abstract_inverted_index.(DCNv4),	5
abstract_inverted_index.building	178
abstract_inverted_index.combined	163
abstract_inverted_index.compared	66
abstract_inverted_index.designed	12
abstract_inverted_index.increase	72, 147
abstract_inverted_index.instance	96
abstract_inverted_index.minimize	53
abstract_inverted_index.notably,	101
abstract_inverted_index.operator	11
abstract_inverted_index.property	43
abstract_inverted_index.removing	33
abstract_inverted_index.semantic	98
abstract_inverted_index.spectrum	16
abstract_inverted_index.speedup.	57
abstract_inverted_index.achieving	78
abstract_inverted_index.addresses	21
abstract_inverted_index.baseline,	119
abstract_inverted_index.diffusion	114
abstract_inverted_index.effective	10
abstract_inverted_index.efficient	8
abstract_inverted_index.including	93
abstract_inverted_index.introduce	1
abstract_inverted_index.potential	174
abstract_inverted_index.practical	128
abstract_inverted_index.redundant	54
abstract_inverted_index.replacing	130
abstract_inverted_index.Deformable	2
abstract_inverted_index.efficiency	160
abstract_inverted_index.expressive	45
abstract_inverted_index.generative	107, 125
abstract_inverted_index.integrated	105
abstract_inverted_index.operations	55
abstract_inverted_index.optimizing	49
abstract_inverted_index.processing	74
abstract_inverted_index.Convolution	3
abstract_inverted_index.InternImage	136
abstract_inverted_index.aggregation	38
abstract_inverted_index.convergence	65
abstract_inverted_index.exceptional	88
abstract_inverted_index.generation.	103
abstract_inverted_index.improvement	151
abstract_inverted_index.limitations	23
abstract_inverted_index.outperforms	117
abstract_inverted_index.performance	89, 150, 167
abstract_inverted_index.possibility	122
abstract_inverted_index.substantial	71
abstract_inverted_index.advancements	156
abstract_inverted_index.demonstrates	87
abstract_inverted_index.foundational	177
abstract_inverted_index.improvements	59
abstract_inverted_index.predecessor,	26
abstract_inverted_index.underscoring	120
abstract_inverted_index.applications,	129
abstract_inverted_index.applications.	19
abstract_inverted_index.enhancements:	31
abstract_inverted_index.normalization	35
abstract_inverted_index.segmentation,	99
abstract_inverted_index.significantly	63
abstract_inverted_index.modifications.	154
abstract_inverted_index.classification,	95
abstract_inverted_index.FlashInternImage	140
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	13
citation_normalized_percentile