Face Expression Recognition via transformer-based classification models
| dc.contributor.author | Arslanoğlu, Muhammed Cihad | |
| dc.contributor.author | Acar, Hüseyin | |
| dc.contributor.author | Albayrak, Abdülkadir | |
| dc.date.accessioned | 2025-03-08T18:25:50Z | |
| dc.date.available | 2025-03-08T18:25:50Z | |
| dc.date.issued | 2024 | |
| dc.department | Dicle Üniversitesi | |
| dc.description.abstract | Facial Expression Recognition (FER) tasks have widely studied in the literature since it has many applications. Fast development of technology in deep learning computer vision algorithms, especially, transformer-based classification models, makes it hard to select most appropriate models. Using complex model may increase accuracy performance but decreasing inference time which is a crucial in near real-time applications. On the other hand, small models may not give desired results. In this study, we aimed to examine performance of 5 different relatively small transformer-based image classification algorithms for FER tasks. We used vanilla ViT, PiT, Swin, DeiT, and CrossViT with considering their trainable parameter size and architectures. Each model has 20-30M trainable parameters which means relatively small. Moreover, each model has different architectures. As an illustration, CrossViT focuses on image using multi-scale patches and PiT model introduces convolution layers and pooling techniques to vanilla ViT model. We obtained all results for widely used FER datasets: CK+ and KDEF. We observed that, PiT model achieves the best accuracy scores 0.9513 and 0.9090 for CK+ and KDEF datasets, respectively | |
| dc.identifier.doi | 10.17694/bajece.1486140 | |
| dc.identifier.endpage | 223 | |
| dc.identifier.issn | 2147-284X | |
| dc.identifier.issn | 2147-284X | |
| dc.identifier.issue | 3 | |
| dc.identifier.startpage | 214 | |
| dc.identifier.uri | https://doi.org/10.17694/bajece.1486140 | |
| dc.identifier.uri | https://hdl.handle.net/11468/30425 | |
| dc.identifier.volume | 12 | |
| dc.language.iso | en | |
| dc.publisher | MUSA YILMAZ | |
| dc.relation.ispartof | Balkan Journal of Electrical and Computer Engineering | |
| dc.relation.publicationcategory | Makale - Ulusal Hakemli Dergi - Kurum Öğretim Elemanı | |
| dc.rights | info:eu-repo/semantics/openAccess | |
| dc.snmz | KA_DergiPark_21250205 | |
| dc.subject | FER | |
| dc.subject | Transformers | |
| dc.subject | ViT | |
| dc.subject | Classification | |
| dc.title | Face Expression Recognition via transformer-based classification models | |
| dc.type | Article |
