Improved vision transformer for sports image classification via transfer learning and knowledge distillation
- 1School of Physical Education, Jining University, Qufu, China
- 2Faculty of Educational Sciences and Technology, Universiti Teknologi Malaysia, Skudai, Malaysia
- 3Institute of Exercise Training and Sport Informatics, German Sport University Cologne, Cologne, Germany
Abstract
To address the challenges posed by complex visual characteristics in sports image classification, this study proposes an enhanced Vision Transformer (ViT) model built upon the Native Resolution ViT (NaViT) architecture. The principal innovations include replacing conventional single-layer feedforward networks with parallel feedforward networks and introducing an attention-guided distillation framework with input-dependent attention-weighted feature fusion, which adaptively regulates the contribution of frozen distillation features to the prediction pathway. The proposed approach leverages ImageNet-pretrained weights and performs domain-specific fine-tuning on sports datasets. Experimental results demonstrate substantial performance gains. On the multi-sport classification dataset, the proposed model achieves an accuracy of 0.9752 and a Macro-F1 score of 0.9753, outperforming the evaluated CNN benchmarks and ViT variants. For soccer match event classification, it achieves an accuracy of 0.9863 and a Macro-F1 score of 0.9718, while ranking first in the TOPSIS evaluation. These results demonstrate the effectiveness of integrating parallel feedforward processing, transfer learning, and distillation-guided feature fusion for sports image classification under complex visual conditions.
Related
- An adaptive transformer-based model with entropy-driven knowledge distillation for single-channel EEG REM sleep detection
- A temperature-adaptive logit-standardized knowledge distillation for lightweight network intrusion detection
- Relational Knowledge Distillation Brings DNN Representations Close Enough to Humans to Be Aligned Without Supervision
- A lightweight privacy-aware federated vision transformer for tomato leaf disease classification
- MangoSwin-attention: a swin transformer with attention pooling for generalizable mango leaf disease classification
BibTeX
@article{li2026improved,
title = {{Improved vision transformer for sports image classification via transfer learning and knowledge distillation}},
author = {Chao Li and Daniel Memmert and Qiang Liu},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0337744},
url = {https://doi.org/10.1371/journal.pone.0337744},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Improved vision transformer for sports image classification via transfer learning and knowledge distillation
AU - Chao Li
AU - Daniel Memmert
AU - Qiang Liu
PY - 2026
DA - 2026/10/16
JO - PLOS One
DO - 10.1371/journal.pone.0337744
UR - https://doi.org/10.1371/journal.pone.0337744
AB - To address the challenges posed by complex visual characteristics in sports image classification, this study proposes an enhanced Vision Transformer (ViT) model built upon the Native Resolution ViT (NaViT) architecture. The principal innovations include replacing conventional single-layer feedforward networks with parallel feedforward networks and introducing an attention-guided distillation framework with input-dependent attention-weighted feature fusion, which adaptively regulates the contribution of frozen distillation features to the prediction pathway. The proposed approach leverages ImageNet-pretrained weights and performs domain-specific fine-tuning on sports datasets. Experimental results demonstrate substantial performance gains. On the multi-sport classification dataset, the proposed model achieves an accuracy of 0.9752 and a Macro-F1 score of 0.9753, outperforming the evaluated CNN benchmarks and ViT variants. For soccer match event classification, it achieves an accuracy of 0.9863 and a Macro-F1 score of 0.9718, while ranking first in the TOPSIS evaluation. These results demonstrate the effectiveness of integrating parallel feedforward processing, transfer learning, and distillation-guided feature fusion for sports image classification under complex visual conditions.
SN - 1932-6203
ER -