A Comprehensive Overview of Deep Learning Models for Object Detection from Videos/Images

2026-01-21

International Journal of Artificial Intelligence and Soft Computing

AI Summary

This study addresses the challenges of robustness and real-time performance in object detection for video and image surveillance under dynamic environments, occlusions, and varying illumination. It presents a systematic review of deep learning–based approaches, offering a novel taxonomy along three dimensions: core architectures, data processing strategies, and surveillance-specific challenges. The work critically examines the roles of CNN-based detectors and generative models—particularly GANs—in tasks such as frame reconstruction, occlusion mitigation, and illumination normalization, alongside mechanisms for temporal information fusion. Through a comprehensive evaluation of prevailing models, benchmark datasets, and performance metrics, the paper delineates the current efficacy boundaries of semantic object detection and identifies promising future directions, including low-latency inference, efficient modeling, and joint spatiotemporal learning.

Technology Category

• Computer Vision: Object Detection & Categorization
• Machine Learning: Deep Generative Models & Autoencoders
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Object detection in video and image surveillance is a well-established yet rapidly evolving task, strongly influenced by recent deep learning advancements. This review summarises modern techniques by examining architectural innovations, generative model integration, and the use of temporal information to enhance robustness and accuracy. Unlike earlier surveys, it classifies methods based on core architectures, data processing strategies, and surveillance specific challenges such as dynamic environments, occlusions, lighting variations, and real-time requirements. The primary goal is to evaluate the current effectiveness of semantic object detection, while secondary aims include analysing deep learning models and their practical applications. The review covers CNN-based detectors, GAN-assisted approaches, and temporal fusion methods, highlighting how generative models support tasks such as reconstructing missing frames, reducing occlusions, and normalising illumination. It also outlines preprocessing pipelines, feature extraction progress, benchmarking datasets, and comparative evaluations. Finally, emerging trends in low-latency, efficient, and spatiotemporal learning approaches are identified for future research.

Problem

• object detection
• video surveillance
• deep learning
• occlusions
• real-time requirements

Innovation

• generative models
• temporal fusion
• occlusion handling
• real-time object detection
• spatiotemporal learning