20.500.12556/RUL-181219 FACES facial analysis with compressed efficient systems Due to their promising performance vision transformers are increasingly being incorporated into various biometric solutions, mainly in the domain of face analysis. However, their size and computational expense remain the biggest challenge when it comes to their full utilization and there is a high demand for optimization of these models. In this paper we propose a novel pruning technique for face analysis vision transformers aimed at reducing their memory and computational cost. The method uses existing transformer parameters as importance scores, which allows for a simple one-shot pruning and retraining approach. By testing the method on the SWINFace transformer for both verification and attribute recognition tasks, we show that the models compressed up to 50% sparsity level maintain the performance or even outperform the original model, while also outperforming state-of-the-art vision transformer pruning methods and showing versatility for different face analysis tasks. face analysis neural network compression pruning vision transformers analiza obraza kompresija nevronskih mrež obrezovanje vidni transformatorji ViT true false true Angleški jezik Slovenski jezik Članek v reviji 2026-03-27 09:55:30 2026-03-27 09:55:31 2026-03-30 04:27:29 0000-00-00 00:00:00 2026 0 0 5 str. no. Vol. 2026 0000-00-00 Zaloznikova Objavljeno apc 0000-00-00 0000-00-00 0000-00-00 004.93:57.087.1 2405-9595 10.1016/j.icte.2026.02.008 273178115 RAZ_Lajic_Romanela_2026.pdf RAZ_Lajic_Romanela_2026.pdf 1 CBC5435D9236009F9C23A1BE15DBE26E 0ac923d21c008831720bef463efcdc6b96e61b37719fccb1856f419e7d2fcfe2 202d2daf-29bb-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=231447 https://www.sciencedirect.com/science/article/pii/S2405959526000299 1 93ed9ba4-29ba-11f1-b0ab-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=231446 Fakulteta za računalništvo in informatiko 0 0 0