Aug 2026· Sri Lankan Journal of Technology· 0 citations
TL;DR
This study systematically compares seven pre-trained feature extractors across three architectural families, convolutional neural networks (CNNs), Vision Transformers (ViTs), and self-supervised models to provide practical guidance on model selection for downstream deep learning tasks.
Abstract
The rapid growth of artificial intelligence and deep learning has revolutionized diverse domains, including computer vision where transfer learning through pre-trained models has become a fundamental technique for enhancing performance of models and reducing computational costs. Although numerous pre-trained deep learning models have been widely adopted, their effectiveness varies significantly across application domains and datasets. The lack of comprehensive comparative studies makes it challenging for researchers to identify the most suitable models for achieving optimal performance in specific tasks. This study systematically compares seven pre-trained feature extractors across three architectural families, convolutional neural networks (CNNs), Vision Transformers (ViTs), and self-supervised models to provide practical guidance on model selection for downstream deep learning tasks. These models were evaluated on five benchmark datasets. Features were extracted with frozen weights and evaluated using linear probing, k-nearest neighbor retrieval, and 5-shot classification. ConvNeXt-B achieved the highest mean linear probe accuracy (90.40%), while DINOv2-S produced the best feature geometry for retrieval tasks (87.64%). CLIP-ViT-B/32 demonstrated the strongest cross-domain transfer, leading on texture recognition and satellite imagery few-shot classification. Older CNN architectures lagged significantly behind modern models by approximately 7%. Overall, this study simultaneously evaluates multiple feature extractors across multiple visual domains and evaluation protocols.
The results demonstrate that transfer learning significantly improves convergence speed, generalization, and computational efficiency, making it a promising approach for AI applications across domains such as healthcare, NLP, and autonomous systems.
Abdul Sttar Ismail Wdaa, Iraq Ali Hussein, A. Ahmed· Future Technology· 0 citations
Experimental findings show that pretrained models outperform those trained from scratch in terms of accuracy, convergence speed, and robustness, and a unified framework is proposed to integrate both processes in a deep learning pipeline.
Kwame Nkosi· International Journal of App...· 0 citations
Tremendous numbers of images and videos are generated throughout the life cycle of construction projects, offering rich visual data that, if analysed effectively, can support automation in progress monitoring, safety management, quality control, and productivity tracking. Within this context, image classification plays a foundational role: it is the supervised task of mapping an image to one of K semantic categories by learning from labelled examples, requiring the extraction of discriminative visual structure and the production of calibrated class predictions. Despite major advances from handcrafted feature pipelines to Convolutional Neural Networks (CNNs), fair comparison across methods remains challenging due to inconsistent preprocessing, dataset splits, training budgets, and reporting standards. This paper introduces a unified and fully reproducible benchmarking framework that standardises preprocessing, enforces fixed train–test splits, aligns training configurations, and evaluates models using a consistent multi-metric suite. Under identical experimental conditions, we benchmark four paradigm families: classical machine learning (ML) models, handcrafted feature-based methods (Histogram of Oriented Gradients (HOG), Local Binary Patterns (LBP), Bag-of-Visual-Words (BoVW) [Formula: see text] Support Vector Machine (SVM)), and CNNs (LeNet-5, ResNet-18), across three datasets (MNIST, Fashion-MNIST, CIFAR-10). We formalise image classification mathematically as a supervised learning problem and assess performance using accuracy, macro-averaged [Formula: see text]1-score, and macro-averaged ROC–AUC, complemented by learning-curve and robustness analyses. Results show that ResNet-18 achieves the strongest performance on all datasets (98.95%, 90.27%, 72.90%), SVM with a Radial Basis Function (RBF) kernel remains competitive on grayscale benchmarks, and hand-crafted pipelines degrade sharply on CIFAR-10. ROC–AUC analysis further highlights calibration–accuracy discrepancies, such as Naïve Bayes achieving high AUC but modest accuracy on MNIST. These findings emphasise the importance of task-aware model selection and demonstrate that compute-aligned, multi-metric benchmarking is essential for fair and reproducible cross-paradigm evaluation. All CNN results are reported under a 50-epoch training budget, which represents a lower bound on converged performance; future extensions will incorporate domain-specific datasets and pre-trained architectures to broaden practical applicability.
Yasir Hamid, N. Ababneh, A. Abasi et al.· Journal of Information &...· 0 citations
Image classification is a major area in computer vision, driven by rapid advances in deep learning. Over the last decade, convolutional neural networks (CNNs) and their variants have achieved high performance in applications such as medical diagnosis, autonomous driving, industrial inspection, remote sensing, and biometrics. However, choosing the right model remains challenging due to trade-offs between accuracy, computational cost, efficiency, and robustness. This paper presents a comparative study of different deep learning architectures, including classical CNNs, deep hierarchical models, residual and dense networks, and compound-scaled architectures. Using a common evaluation framework and standard datasets, the study analyzes performance based on key design factors such as depth, width, receptive field, skip connections, and normalization. Theoretical concepts like convolution operations, residual learning, and optimization are also discussed.The results show that deeper networks provide better representation, while residual connections and compound scaling improve training stability and efficiency. Lightweight models perform well in resource-limited and real-time environments. Overall, the study offers practical guidance for selecting suitable architectures and highlights future research areas such as neural architecture search, self-supervised learning, and efficient model deployment.
Riyaz Mohammed· International Journal of App...· 0 citations
Results consistently show that KD most effectively benefits student models of intermediate complexity across all three architectures, with the distilled FCN student reducing parameters by a factor of 38, the distilled Inception student achieving nearly the same performance as the teacher with 42% fewer parameters and the distilled ConvTran student with 2 attention heads showing the most significant improvement through distillation.
Javidan Abdullayev, M. Devanne, Jonathan Weber et al.· Knowledge and Information Sy...· 0 citations
Convolutional Neural Networks (CNNs) have significantly changed image classification over the years by allowing computers to learn features directly from raw pixel data. However, deciding between building a customised model and using a pre-trained one can be a difficult task, especially when working with small datasets. In this study, we compare a custom CNN with three pre-trained models—VGG16, ResNet50, and MobileNetV2—on the CIFAR-10 dataset, which comprises 60,000 colour images (32×32 pixels) across 10 categories. We measured model performance using accuracy, precision, recall, F1-score, and training time. The results show that pre-trained models performed much better than the customised model. ResNet50 had the highest accuracy at 92.4%. However, MobileNetV2 gave the best mix of speed (1,800 seconds to train) and accuracy (90.2%). The custom CNN reached 82.3% accuracy, used less memory, and did not need image resizing. These results offer clear benchmarks for choosing models in the face of limited resources. They also demonstrate that transfer learning can achieve strong performance, while showing that custom CNNs remain useful for learning and simple tasks.
Oluwadamilare (Asabia) Joseph Omoniyi, Omotosho Olawale Jacob, Ajaegbu Chigozirim et al.· International journal of res...· 0 citations