Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2407.08113 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909250562818048 |
|---|---|
| author | Son, Byunggwan Oh, Youngmin Baek, Donghyeon Ham, Bumsub |
| author_facet | Son, Byunggwan Oh, Youngmin Baek, Donghyeon Ham, Bumsub |
| contents | Dataset distillation synthesizes a small set of images from a large-scale real dataset such that synthetic and real images share similar behavioral properties (e.g, distributions of gradients or features) during a training process. Through extensive analyses on current methods and real datasets, together with empirical observations, we provide in this paper two important things to share for dataset distillation. First, object parts that appear on one side of a real image are highly likely to appear on the opposite side of another image within a dataset, which we call the bilateral equivalence. Second, the bilateral equivalence enforces synthetic images to duplicate discriminative parts of objects on both the left and right sides of the images, limiting the recognition of subtle differences between objects. To address this problem, we introduce a surprisingly simple yet effective technique for dataset distillation, dubbed FYI, that enables distilling rich semantics of real images into synthetic ones. To this end, FYI embeds a horizontal flipping technique into distillation processes, mitigating the influence of the bilateral equivalence, while capturing more details of objects. Experiments on CIFAR-10/100, Tiny-ImageNet, and ImageNet demonstrate that FYI can be seamlessly integrated into several state-of-the-art methods, without modifying training objectives and network architectures, and it improves the performance remarkably. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_08113 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FYI: Flip Your Images for Dataset Distillation Son, Byunggwan Oh, Youngmin Baek, Donghyeon Ham, Bumsub Computer Vision and Pattern Recognition Dataset distillation synthesizes a small set of images from a large-scale real dataset such that synthetic and real images share similar behavioral properties (e.g, distributions of gradients or features) during a training process. Through extensive analyses on current methods and real datasets, together with empirical observations, we provide in this paper two important things to share for dataset distillation. First, object parts that appear on one side of a real image are highly likely to appear on the opposite side of another image within a dataset, which we call the bilateral equivalence. Second, the bilateral equivalence enforces synthetic images to duplicate discriminative parts of objects on both the left and right sides of the images, limiting the recognition of subtle differences between objects. To address this problem, we introduce a surprisingly simple yet effective technique for dataset distillation, dubbed FYI, that enables distilling rich semantics of real images into synthetic ones. To this end, FYI embeds a horizontal flipping technique into distillation processes, mitigating the influence of the bilateral equivalence, while capturing more details of objects. Experiments on CIFAR-10/100, Tiny-ImageNet, and ImageNet demonstrate that FYI can be seamlessly integrated into several state-of-the-art methods, without modifying training objectives and network architectures, and it improves the performance remarkably. |
| title | FYI: Flip Your Images for Dataset Distillation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.08113 |