Diffusion-Enhanced Test-time Adaptation with Text and Image Augmentation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915080408399872 |
|---|---|
| author | Feng, Chun-Mei He, Yuanyang Zou, Jian Khan, Salman Xiong, Huan Li, Zhen Zuo, Wangmeng Goh, Rick Siow Mong Liu, Yong |
| author_facet | Feng, Chun-Mei He, Yuanyang Zou, Jian Khan, Salman Xiong, Huan Li, Zhen Zuo, Wangmeng Goh, Rick Siow Mong Liu, Yong |
| contents | Existing test-time prompt tuning (TPT) methods focus on single-modality data, primarily enhancing images and using confidence ratings to filter out inaccurate images. However, while image generation models can produce visually diverse images, single-modality data enhancement techniques still fail to capture the comprehensive knowledge provided by different modalities. Additionally, we note that the performance of TPT-based methods drops significantly when the number of augmented images is limited, which is not unusual given the computational expense of generative augmentation. To address these issues, we introduce IT3A, a novel test-time adaptation method that utilizes a pre-trained generative model for multi-modal augmentation of each test sample from unknown new domains. By combining augmented data from pre-trained vision and language models, we enhance the ability of the model to adapt to unknown new test data. Additionally, to ensure that key semantics are accurately retained when generating various visual and text enhancements, we employ cosine similarity filtering between the logits of the enhanced images and text with the original test data. This process allows us to filter out some spurious augmentation and inadequate combinations. To leverage the diverse enhancements provided by the generation model across different modals, we have replaced prompt tuning with an adapter for greater flexibility in utilizing text templates. Our experiments on the test datasets with distribution shifts and domain gaps show that in a zero-shot setting, IT3A outperforms state-of-the-art test-time prompt tuning methods with a 5.50% increase in accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_09706 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Diffusion-Enhanced Test-time Adaptation with Text and Image Augmentation Feng, Chun-Mei He, Yuanyang Zou, Jian Khan, Salman Xiong, Huan Li, Zhen Zuo, Wangmeng Goh, Rick Siow Mong Liu, Yong Computer Vision and Pattern Recognition Existing test-time prompt tuning (TPT) methods focus on single-modality data, primarily enhancing images and using confidence ratings to filter out inaccurate images. However, while image generation models can produce visually diverse images, single-modality data enhancement techniques still fail to capture the comprehensive knowledge provided by different modalities. Additionally, we note that the performance of TPT-based methods drops significantly when the number of augmented images is limited, which is not unusual given the computational expense of generative augmentation. To address these issues, we introduce IT3A, a novel test-time adaptation method that utilizes a pre-trained generative model for multi-modal augmentation of each test sample from unknown new domains. By combining augmented data from pre-trained vision and language models, we enhance the ability of the model to adapt to unknown new test data. Additionally, to ensure that key semantics are accurately retained when generating various visual and text enhancements, we employ cosine similarity filtering between the logits of the enhanced images and text with the original test data. This process allows us to filter out some spurious augmentation and inadequate combinations. To leverage the diverse enhancements provided by the generation model across different modals, we have replaced prompt tuning with an adapter for greater flexibility in utilizing text templates. Our experiments on the test datasets with distribution shifts and domain gaps show that in a zero-shot setting, IT3A outperforms state-of-the-art test-time prompt tuning methods with a 5.50% increase in accuracy. |
| title | Diffusion-Enhanced Test-time Adaptation with Text and Image Augmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.09706 |