Ali YILMAZ

Mr. Ali YILMAZ

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 March 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
AYAli YILMAZ
Ali YILMAZ
YYYüksel Yurtay
Yüksel Yurtay
NYNilüfer Yurtay
Nilüfer Yurtay
AYOLO introduces a novel fusion architecture that integrates unsupervised learning techniques with Vision Transformers, leveraging the YOLO series models as its foundation. This innovation enables the effective utilization of rich, unlabeled data, establishing a new pretraining methodology tailored to YOLO architectures. On a custom dataset comprising 80 images of poppy plants, AYOLO achieved a remarkable Average Precision (AP) of 38.7% while maintaining a high rendering speed of 239 FPS (Frames Per Second) on a Tesla K80 GPU. Real-time performance is demonstrated by achieving 239 FPS, and feature fusion optimally combines spatial and semantic information across scales. This performance surpasses the previous state-of-the-art YOLO v6-3.0 by +2.2% AP while retaining comparable speed. AYOLO exemplifies the potential of integrating advanced information fusion techniques with supervised pretraining, significantly enhancing precision and efficiency for object detection models optimized for small, specialized datasets.

13 September 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
AYAli YILMAZ
Ali YILMAZ
YYYüksel Yurtay
Yüksel Yurtay
NYNilüfer Yurtay
Nilüfer Yurtay
AYOLO differentiates itself with a new fusion architecture proposal that leverages the strengths of unsupervised learning and integrates the Vision Transformer approach, taking the YOLO series models as a reference. This innovation enables the model to effectively use rich, unlabeled data, providing a new example in pre-training methodology for YOLO architectures. On the benchmark COCO val2017 dataset, AYOLO demonstrates its superiority by achieving an impressive 38.7 % AP while maintaining an outstanding rendering speed of 239 FPS (Frame Per Second) on the Tesla K80 GPU. This performance outperforms the previous state-of-the-art YOLO v6-3.0 by a significant margin of +2.2 % AP while maintaining comparable FPS. AYOLO is presented as a demonstration of the potential of integrating complex information fusion techniques with supervised pretraining in improving the precision and speed of object detection models.

25 October 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
AYAli YILMAZ
Ali YILMAZ
YYYüksel Yurtay
Yüksel Yurtay
NYNilüfer Yurtay
Nilüfer Yurtay
AYOLO differentiates itself with a new fusion architecture proposal that leverages the strengths of unsupervised learning and integrates the Vision Transformer approach, taking the YOLO series models as a reference. This innovation enables the model to effectively use rich, unlabeled data, providing a new example in pre-training methodology for YOLO architectures. On the benchmark COCO val2017 dataset, AYOLO demonstrates its superiority by achieving an impressive 38.7 % AP while maintaining an outstanding rendering speed of 239 FPS (Frame Per Second) on the Tesla K80 GPU. This performance outperforms the previous state-of-the-art YOLO v6-3.0 by a significant margin of +2.2 % AP while maintaining comparable FPS. AYOLO is presented as a demonstration of the potential of integrating complex information fusion techniques with supervised pretraining in improving the precision and speed of object detection models.