|
|
|
|
| LEADER |
01000caa a22002652c 4500 |
| 001 |
NLM392639793 |
| 003 |
DE-627 |
| 005 |
20250923233015.0 |
| 007 |
cr uuu---uuuuu |
| 008 |
250917s2025 xx |||||o 00| ||eng c |
| 024 |
7 |
|
|a 10.1109/TIP.2025.3607612
|2 doi
|
| 028 |
5 |
2 |
|a pubmed25n1578.xml
|
| 035 |
|
|
|a (DE-627)NLM392639793
|
| 035 |
|
|
|a (NLM)40953421
|
| 040 |
|
|
|a DE-627
|b ger
|c DE-627
|e rakwb
|
| 041 |
|
|
|a eng
|
| 100 |
1 |
|
|a Tang, Chenwei
|e verfasserin
|4 aut
|
| 245 |
1 |
0 |
|a Rethinking Generalized Zero-Shot Learning
|b A Synthesized Per-Instance Attribute Perspective
|
| 264 |
|
1 |
|c 2025
|
| 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 Revised 22.09.2025
|
| 500 |
|
|
|a published: Print
|
| 500 |
|
|
|a Citation Status PubMed-not-MEDLINE
|
| 520 |
|
|
|a Generalized zero-shot learning (GZSL) shows great potential for improving generalization to unseen classes in real-world scenarios. However, most GZSL methods depend on benchmark datasets with per-class attribute annotations, which creates a large semantic gap and worsens the domain shift problem in the visual-semantic space. To address these challenges, instance-level attributes offer an intuitive solution, but they require expensive manual annotation. In this paper, we propose a simple yet effective approach called per-instance attribute synthesis (PIAS) to generate diverse semantic representations for each instance. Our method first uses the Vision Transformer (ViT) model to extract visual features and then generates per-instance attributes. The patch splitting, positional embedding, and multi-head self-attention mechanisms in ViT improve the discriminability of both visual and semantic representations. Next, we define the generated attributes of class-average images as class anchor points. These anchor points are calibrated in the semantic space by minimizing the cosine similarity between the anchor points and per-class attribute annotations. Finally, we improve the diversity of generated per-instance attributes by aligning the topological structure between per-class attribute annotations and synthesized per-instance attributes with that between class-average visual features and per-instance visual features. We conduct comprehensive experiments on three challenging ZSL datasets: AWA2, CUB, and SUN. The results show that PIAS significantly outperforms state-of-the-art methods under both ZSL and GZSL settings. We further demonstrate the generalization ability of PIAS by applying it to attribute-based zero-shot image retrieval tasks
|
| 650 |
|
4 |
|a Journal Article
|
| 700 |
1 |
|
|a Wang, Ying
|e verfasserin
|4 aut
|
| 700 |
1 |
|
|a Xie, Wei
|e verfasserin
|4 aut
|
| 700 |
1 |
|
|a Zhang, Qianjun
|e verfasserin
|4 aut
|
| 700 |
1 |
|
|a Xiao, Rong
|e verfasserin
|4 aut
|
| 700 |
1 |
|
|a He, Zhenan
|e verfasserin
|4 aut
|
| 700 |
1 |
|
|a Lv, Jiancheng
|e verfasserin
|4 aut
|
| 773 |
0 |
8 |
|i Enthalten in
|t IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
|d 1992
|g 34(2025) vom: 12., Seite 5847-5859
|w (DE-627)NLM09821456X
|x 1941-0042
|7 nnas
|
| 773 |
1 |
8 |
|g volume:34
|g year:2025
|g day:12
|g pages:5847-5859
|
| 856 |
4 |
0 |
|u http://dx.doi.org/10.1109/TIP.2025.3607612
|3 Volltext
|
| 912 |
|
|
|a GBV_USEFLAG_A
|
| 912 |
|
|
|a SYSFLAG_A
|
| 912 |
|
|
|a GBV_NLM
|
| 912 |
|
|
|a GBV_ILN_350
|
| 951 |
|
|
|a AR
|
| 952 |
|
|
|d 34
|j 2025
|b 12
|h 5847-5859
|