Computer Vision in the Deep Learning Era: A Survey of Architectures, Datasets, and Applications

From Convolution to Attention and Beyond: A Systematic Review of Modern Vision Architectures

A critical review of computer vision, illustrating how architectural design, learning paradigms, and evaluation practices have co-evolved over time to facilitate more flexible and scalable systems, and outlining new research directions.

A Comparative Study of Deep Learning Architectures for Image Classification

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.

Texture Representations in Deep Vision Models: Comparing CNNs, Vision Transformers, and Human Perception

In computational vision science, Convolutional Neural Networks (CNNs) have emerged as a popular model of biological vision because of the alignment they can exhibit with neural and behavioral data in humans and animals. However, it remains unclear to what extent this alignment persists for visual tasks that extend beyond the canonical object recognition paradigm based on well defined semantic content. In this study, we diverge from the common object-centric view by focusing on another aspect of vision: texture perception. We consider textures of different complexity generated with three different algorithms from the same source images. Using a rank-based statistic, we quantify the information encoded in the internal representations of a CNN and three Vision Transformers (ViTs), and we compare the similarity of these representations to those inferred from human psychophysics data. We find that the representation of textures is aligned in different ViTs, but not between the ViTs and the CNN; that ViTs form similar representations for textures of different complexity; that human performance in recognizing textures can be better predicted from ViTs representations rather than CNN representations. Taken together, these results suggest that ViTs may capture more faithfully than CNNs how texture patterns are visually processed by humans, and that the representations of texture stimuli in computational models may be driven by the network architecture.

DeepVision: A Hybrid Deepfake Detection Framework Using Deep Learning Approaches

Over the past decade, rapid progress in artificial intelligence (AI), machine learning, and deep learning has introduced sophisticated techniques for multimedia manipulation. Although such technologies have legitimate applications in entertainment and education, malicious actors increasingly exploit them for disinformation campaigns, political propaganda, identity fraud, and targeted harassment. High-quality synthetic videos and images commonly known as deepfakes pose a growing threat to digital security and public trust. This paper introduces DeepVision, a hybrid deepfake detection framework that fuses EfficientNet-B0 with a Vision Transformer (ViTB/16) to exploit both local texture features and global spatial dependencies simultaneously. The EfficientNet-B0 branch extracts fine-grained local texture and manipulation artefacts, while the Vision Transformer captures long range contextual relationships across facial regions using multi-head self-attention. The model is trained on a combined dataset derived from FaceForensics++ (FF++) and the DeepFake Detection Challenge (DFDC), comprising 120,000 labeled face images. Model performance is evaluated using accuracy, precision, recall, F1- score, confusion matrix, and ROC-AUC metrics. Experimental results demonstrate strong classification performance, achieving 98% accuracy and an AUC of 0.9973 on the combined dataset, representing competitive performance relative to recent state-of-the-art studies. The proposed framework supports both image-based and video-based deepfake detection and is suitable for real-world deployment in digital forensics and media authentication applications.

A Comparative Analysis of Feature-Based and Deep Learning Approaches for Image Classification

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.

Performance Comparison of ALEXNET, VGG16, and ResNet on CIFAR‑10 Image Classification