Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2507.01026 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915368303329280 |
|---|---|
| author | Shohag, Md Shakil Ahamed Wu, Q. M. Jonathan Pourpanah, Farhad |
| author_facet | Shohag, Md Shakil Ahamed Wu, Q. M. Jonathan Pourpanah, Farhad |
| contents | Generative zero-shot learning (ZSL) methods typically synthesize visual features for unseen classes using predefined semantic attributes, followed by training a fully supervised classification model. While effective, these methods require substantial computational resources and extensive synthetic data, thereby relaxing the original ZSL assumptions. In this paper, we propose FSIGenZ, a few-shot-inspired generative ZSL framework that reduces reliance on large-scale feature synthesis. Our key insight is that class-level attributes exhibit instance-level variability, i.e., some attributes may be absent or partially visible, yet conventional ZSL methods treat them as uniformly present. To address this, we introduce Model-Specific Attribute Scoring (MSAS), which dynamically re-scores class attributes based on model-specific optimization to approximate instance-level variability without access to unseen data. We further estimate group-level prototypes as clusters of instances based on MSAS-adjusted attribute scores, which serve as representative synthetic features for each unseen class. To mitigate the resulting data imbalance, we introduce a Dual-Purpose Semantic Regularization (DPSR) strategy while training a semantic-aware contrastive classifier (SCC) using these prototypes. Experiments on SUN, AwA2, and CUB benchmarks demonstrate that FSIGenZ achieves competitive performance using far fewer synthetic features. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_01026 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Few-Shot Inspired Generative Zero-Shot Learning Shohag, Md Shakil Ahamed Wu, Q. M. Jonathan Pourpanah, Farhad Machine Learning Generative zero-shot learning (ZSL) methods typically synthesize visual features for unseen classes using predefined semantic attributes, followed by training a fully supervised classification model. While effective, these methods require substantial computational resources and extensive synthetic data, thereby relaxing the original ZSL assumptions. In this paper, we propose FSIGenZ, a few-shot-inspired generative ZSL framework that reduces reliance on large-scale feature synthesis. Our key insight is that class-level attributes exhibit instance-level variability, i.e., some attributes may be absent or partially visible, yet conventional ZSL methods treat them as uniformly present. To address this, we introduce Model-Specific Attribute Scoring (MSAS), which dynamically re-scores class attributes based on model-specific optimization to approximate instance-level variability without access to unseen data. We further estimate group-level prototypes as clusters of instances based on MSAS-adjusted attribute scores, which serve as representative synthetic features for each unseen class. To mitigate the resulting data imbalance, we introduce a Dual-Purpose Semantic Regularization (DPSR) strategy while training a semantic-aware contrastive classifier (SCC) using these prototypes. Experiments on SUN, AwA2, and CUB benchmarks demonstrate that FSIGenZ achieves competitive performance using far fewer synthetic features. |
| title | Few-Shot Inspired Generative Zero-Shot Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2507.01026 |