Object Detection Revisited: A Systematic Survey from Traditional Vision Pipelines to Transformer-Based Models

Author(s)	Mr. Mushtaq AHMAD DAR
Country	India
Abstract	Object detection aims to localize and recognize objects within images, a challenging task due to variations in scale, occlusion, and scene complexity. Traditional feature-based methods were limited in adaptability, while CNN-based detectors improved performance but often rely on complex multi-stage pipelines. Transformer-based architectures reformulate detection as an end-to-end set prediction problem, leveraging attention mechanisms for global context. This paper systematically surveys object detection techniques from traditional to CNN- and transformer-based models. We analyze architectural choices, performance trade-offs, and real-world applications, using a fuzzy Multi-Criteria Decision Making (MCDM) framework to rank detectors based on speed (FPS) and accuracy (mAP). Open challenges related to efficiency, robustness, and data dependency are discussed, and future research directions including hybrid CNN–transformer models and lightweight architectures are highlighted.
Keywords	Object detection, deep learning, CNN, transformers, DETR, YOLO, fuzzy MCDM, computer vision
Field	Computer > Data / Information
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-05-29
DOI	https://doi.org/10.36948/ijfmr.2026.v08i03.79611

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research