repository
Weighted Alignment-based Multi Source Domain Adaptation for Object Detection
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | 황원준 | - |
| dc.contributor.author | 한준환 | - |
| dc.date.accessioned | 2025-01-25T01:36:11Z | - |
| dc.date.available | 2025-01-25T01:36:11Z | - |
| dc.date.issued | 2023-02 | - |
| dc.identifier.other | 32446 | - |
| dc.identifier.uri | https://dspace.ajou.ac.kr/handle/2018.oak/24707 | - |
| dc.description | 학위논문(석사)--아주대학교 일반대학원 :인공지능학과,2023. 2 | - |
| dc.description.tableofcontents | Ⅰ Introduction 1 <br>Ⅱ. Related Work 4 <br>Ⅲ. Proposed Method 7 <br> A. Framework Overview 7 <br> B. Proposed Method 8 <br>Ⅳ. Experimental Results and Discussion 10 <br> A. Implementation Details 10 <br> B. Datasets 10 <br> C. Comparative Approaches 11 <br> D. Cross Camera Adaptation 11 <br> E. Cross Time Adaptation 14 <br> F. Ablation Study: Domain Shift Aware Adaptation 17 <br> G. Ablation Study: Class Normalization Teaching 18 <br>Ⅴ. Conclusion 19 <br>Reference 20 | - |
| dc.language.iso | eng | - |
| dc.publisher | The Graduate School, Ajou University | - |
| dc.rights | 아주대학교 논문은 저작권에 의해 보호받습니다. | - |
| dc.title | Weighted Alignment-based Multi Source Domain Adaptation for Object Detection | - |
| dc.type | Thesis | - |
| dc.contributor.affiliation | 아주대학교 대학원 | - |
| dc.contributor.alternativeName | Han Joonhwan | - |
| dc.contributor.department | 일반대학원 인공지능학과 | - |
| dc.date.awarded | 2023-02 | - |
| dc.description.degree | Master | - |
| dc.identifier.localId | T000000032446 | - |
| dc.identifier.url | https://dcoll.ajou.ac.kr/dcollection/common/orgView/000000032446 | - |
| dc.subject.keyword | Computer vision | - |
| dc.subject.keyword | Domain Adaptation | - |
| dc.subject.keyword | Multi-Source Domain Adaptation | - |
| dc.description.alternativeAbstract | Deep learning models tends to rely on training datasets, so when new data are entered, performance degradation occurs. Domain adaptation (DA) is a learning technique that reduces the domain shift between a labeled training dataset (the “source dataset”) and an unlabeled target dataset to prevent performance against the target dataset. Although great effort has been achieved in single-source domain adaptation for object detection (DAOD), multi-source domain adaptation for object detection (MSDAOD) has not been studied much yet due to the challenge of considering Domain Shift between Source and Target domains as well as Domain Shift between Source datasets. Recently, previous studies designed Discriminator as many as the number of source datasets to solve each domain shift and conducted alignment for all instances. However, separate alignment makes latent spaces as many as source datasets. So, it is difficult to align all the datasets at once. This motivates us to propose a novel method for multi-source detector adaptation based on multiple binary cross entropy. We build a single latent space that can align the target domain and all the source domains at once. Major cause of performance degradation in DAOD is alignment between data with large domain shift. So, we design the alignment in inverse proportion to domain shift. We experimentally verify the effect of our method on two evaluation protocols. | - |
- Appears in Collections:
- Graduate School of Ajou University > Department of Artificial Intelligence > 3. Theses(Master)
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
-
- License
- STATISTICS
- Total Visit :5,235,105
- Total Download :2,137
- Today View :2,943
