Task-Specific Normalization for Continual Learning of Blind Image Quality Models
In this paper, we present a simple yet effective continual learning method for blind image quality assessment (BIQA) with improved quality prediction accuracy, plasticity-stability trade-off, and task-order/-length robustness. The key step in our approach is to freeze all convolution filters of a pr...
Guardat en:
| Autors principals: | , , , |
|---|---|
| Format: | Article |
| Publicat: |
2021
|
| Matèries: | |
| Accés en línia: | https://arxiv.org/abs/2107.13429 http://arxiv.org/abs/2107.13429 |
| Etiquetes: |
Sense etiquetes, Sigues el primer a etiquetar aquest registre!
|
| _version_ | 1879175377812717568 |
|---|---|
| author | Zhang, Weixia Ma, Kede Zhai, Guangtao Yang, Xiaokang |
| author_browse | Ma, Kede Yang, Xiaokang Zhai, Guangtao Zhang, Weixia |
| author_facet | Zhang, Weixia Ma, Kede Zhai, Guangtao Yang, Xiaokang |
| author_sort | Zhang, Weixia |
| building | arXiv |
| collection | Acceso abierto |
| description | In this paper, we present a simple yet effective continual learning method for blind image quality assessment (BIQA) with improved quality prediction accuracy, plasticity-stability trade-off, and task-order/-length robustness. The key step in our approach is to freeze all convolution filters of a pre-trained deep neural network (DNN) for an explicit promise of stability, and learn task-specific normalization parameters for plasticity. We assign each new IQA dataset (i.e., task) a prediction head, and load the corresponding normalization parameters to produce a quality score. The final quality estimate is computed by black a weighted summation of predictions from all heads with a lightweight $K$-means gating mechanism. Extensive experiments on six IQA datasets demonstrate the advantages of the proposed method in comparison to previous training techniques for BIQA. |
| first_indexed | 2026-10-16T00:28:06Z |
| format | Article |
| fullrecord | <oai_dc:dc xmlns:oai_dc="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/oai_dc/ https://www.openarchives.org/OAI/2.0/oai_dc.xsd"><identifier>arxiv-2107.13429</identifier><datestamp>2024-02-20</datestamp>
<dc:title>Task-Specific Normalization for Continual Learning of Blind Image Quality Models</dc:title>
<dc:creator>Zhang, Weixia</dc:creator>
<dc:creator>Ma, Kede</dc:creator>
<dc:creator>Zhai, Guangtao</dc:creator>
<dc:creator>Yang, Xiaokang</dc:creator>
<dc:subject>Computer Vision and Pattern Recognition</dc:subject>
<dc:description>In this paper, we present a simple yet effective continual learning method for blind image quality assessment (BIQA) with improved quality prediction accuracy, plasticity-stability trade-off, and task-order/-length robustness. The key step in our approach is to freeze all convolution filters of a pre-trained deep neural network (DNN) for an explicit promise of stability, and learn task-specific normalization parameters for plasticity. We assign each new IQA dataset (i.e., task) a prediction head, and load the corresponding normalization parameters to produce a quality score. The final quality estimate is computed by black a weighted summation of predictions from all heads with a lightweight $K$-means gating mechanism. Extensive experiments on six IQA datasets demonstrate the advantages of the proposed method in comparison to previous training techniques for BIQA. </dc:description>
<dc:description>Accepted by IEEE T-IP</dc:description>
<dc:date>2021-07-28</dc:date>
<dc:date>2024-02-19</dc:date>
<dc:type>text</dc:type>
<dc:identifier>https://arxiv.org/abs/2107.13429</dc:identifier>
<dc:identifier>http://arxiv.org/abs/2107.13429</dc:identifier>
</oai_dc:dc> |
| id | arxiv-2107.13429 |
| institution | arXiv |
| last_indexed | 2026-10-16T00:28:06Z |
| publishDate | 2021 |
| publishDateRange | 2021 |
| publishDateSort | 2021 |
| record_format | oai_dc |
| spelling | arxiv-2107.134292024-02-20 Task-Specific Normalization for Continual Learning of Blind Image Quality Models Zhang, Weixia Ma, Kede Zhai, Guangtao Yang, Xiaokang Computer Vision and Pattern Recognition In this paper, we present a simple yet effective continual learning method for blind image quality assessment (BIQA) with improved quality prediction accuracy, plasticity-stability trade-off, and task-order/-length robustness. The key step in our approach is to freeze all convolution filters of a pre-trained deep neural network (DNN) for an explicit promise of stability, and learn task-specific normalization parameters for plasticity. We assign each new IQA dataset (i.e., task) a prediction head, and load the corresponding normalization parameters to produce a quality score. The final quality estimate is computed by black a weighted summation of predictions from all heads with a lightweight $K$-means gating mechanism. Extensive experiments on six IQA datasets demonstrate the advantages of the proposed method in comparison to previous training techniques for BIQA. Accepted by IEEE T-IP 2021-07-28 2024-02-19 text https://arxiv.org/abs/2107.13429 http://arxiv.org/abs/2107.13429 |
| spellingShingle | Computer Vision and Pattern Recognition Zhang, Weixia Ma, Kede Zhai, Guangtao Yang, Xiaokang Task-Specific Normalization for Continual Learning of Blind Image Quality Models |
| title | Task-Specific Normalization for Continual Learning of Blind Image Quality Models |
| title_full | Task-Specific Normalization for Continual Learning of Blind Image Quality Models |
| title_fullStr | Task-Specific Normalization for Continual Learning of Blind Image Quality Models |
| title_full_unstemmed | Task-Specific Normalization for Continual Learning of Blind Image Quality Models |
| title_short | Task-Specific Normalization for Continual Learning of Blind Image Quality Models |
| title_sort | task specific normalization for continual learning of blind image quality models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2107.13429 http://arxiv.org/abs/2107.13429 |
| work_keys_str_mv | AT zhangweixia taskspecificnormalizationforcontinuallearningofblindimagequalitymodels AT makede taskspecificnormalizationforcontinuallearningofblindimagequalitymodels AT zhaiguangtao taskspecificnormalizationforcontinuallearningofblindimagequalitymodels AT yangxiaokang taskspecificnormalizationforcontinuallearningofblindimagequalitymodels |