Skip-layer network with optimization method for domain adaptive detection
In the field of object detection, domain adaptation is one of popular solution to align the distribution between the real scene (target domain) and the training scene (source domain) by adversarial training. However, only global features are applied to the Domain Adaptive Faster R-CNN (DA Faster R-CNN) method. The lack of local features reduces the performance of domain adaptation. Therefore, a novel method for domain adaptive detection called Skip-Layer Network with Optimization (SLNO) method is proposed in this paper. Three improvements are presented in SLNO. Firstly, different level convolutional features are fused by a multi-level features fusion component for domain classifier. Secondly, a multi-layer domain adaptation component is developed to align the image-level and the instance-level distributions simultaneously. Among this component, domain classifiers are used in both image-level and instance-level distributions through the skip layer. Thirdly, the cuckoo search (CS) optimization method is applied to search for the best coefficient of SLNO. As a result, the capability of domain alignment is strengthened. The Cityscapes, Foggy Cityscapes, SIM10K, KITTI data sets are applied to test our proposed novel approach. Consequently, excellent results are achieved by our proposed methods against state-of-the-art object detection methods. The results demonstrate our improvements are effective on domain adaptation detection.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2022 |
---|---|
Erschienen: |
2022 |
Enthalten in: |
Zur Gesamtaufnahme - volume:17 |
---|---|
Enthalten in: |
PloS one - 17(2022), 3 vom: 18., Seite e0263748 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Xu, Qian [VerfasserIn] |
---|
Links: |
---|
Themen: |
---|
Anmerkungen: |
Date Completed 21.04.2022 Date Revised 21.04.2022 published: Electronic-eCollection Citation Status PubMed-not-MEDLINE |
---|
doi: |
10.1371/journal.pone.0263748 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM338291199 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM338291199 | ||
003 | DE-627 | ||
005 | 20231226000336.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231226s2022 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1371/journal.pone.0263748 |2 doi | |
028 | 5 | 2 | |a pubmed24n1127.xml |
035 | |a (DE-627)NLM338291199 | ||
035 | |a (NLM)35298479 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Xu, Qian |e verfasserin |4 aut | |
245 | 1 | 0 | |a Skip-layer network with optimization method for domain adaptive detection |
264 | 1 | |c 2022 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 21.04.2022 | ||
500 | |a Date Revised 21.04.2022 | ||
500 | |a published: Electronic-eCollection | ||
500 | |a Citation Status PubMed-not-MEDLINE | ||
520 | |a In the field of object detection, domain adaptation is one of popular solution to align the distribution between the real scene (target domain) and the training scene (source domain) by adversarial training. However, only global features are applied to the Domain Adaptive Faster R-CNN (DA Faster R-CNN) method. The lack of local features reduces the performance of domain adaptation. Therefore, a novel method for domain adaptive detection called Skip-Layer Network with Optimization (SLNO) method is proposed in this paper. Three improvements are presented in SLNO. Firstly, different level convolutional features are fused by a multi-level features fusion component for domain classifier. Secondly, a multi-layer domain adaptation component is developed to align the image-level and the instance-level distributions simultaneously. Among this component, domain classifiers are used in both image-level and instance-level distributions through the skip layer. Thirdly, the cuckoo search (CS) optimization method is applied to search for the best coefficient of SLNO. As a result, the capability of domain alignment is strengthened. The Cityscapes, Foggy Cityscapes, SIM10K, KITTI data sets are applied to test our proposed novel approach. Consequently, excellent results are achieved by our proposed methods against state-of-the-art object detection methods. The results demonstrate our improvements are effective on domain adaptation detection | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
700 | 1 | |a Li, Ying |e verfasserin |4 aut | |
700 | 1 | |a Wang, Gang |e verfasserin |4 aut | |
700 | 1 | |a Hou, Minghui |e verfasserin |4 aut | |
700 | 1 | |a Zhang, Hao |e verfasserin |4 aut | |
700 | 1 | |a Cai, Hongmin |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t PloS one |d 2006 |g 17(2022), 3 vom: 18., Seite e0263748 |w (DE-627)NLM167327399 |x 1932-6203 |7 nnns |
773 | 1 | 8 | |g volume:17 |g year:2022 |g number:3 |g day:18 |g pages:e0263748 |
856 | 4 | 0 | |u http://dx.doi.org/10.1371/journal.pone.0263748 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 17 |j 2022 |e 3 |b 18 |h e0263748 |