This paper provides a comprehensive survey of object detection over the past 20 years, covering technical evolution, key technologies, and recent state-of-the-art methods. Object detection is a fundamental task in computer vision, aiming to identify objects in images. It has evolved significantly, with early methods relying on handcrafted features and later methods leveraging deep learning. The paper reviews milestone detectors, datasets, metrics, and technical advancements, including the transition from traditional methods to deep learning-based approaches. It highlights the development of two-stage and one-stage detectors, the role of feature pyramids, and the integration of attention mechanisms. The paper also discusses challenges such as object localization, speed, and accuracy, and presents recent advancements like DETR and CenterNet. It covers object detection datasets, evaluation metrics, and technical evolution, including multi-scale detection, context priming, hard negative mining, and loss functions. The paper also addresses speed-up techniques, including feature map sharing, cascaded detection, network pruning, lightweight network design, and numerical acceleration. Overall, the survey provides a detailed overview of the evolution of object detection, emphasizing its technical progress and future directions.This paper provides a comprehensive survey of object detection over the past 20 years, covering technical evolution, key technologies, and recent state-of-the-art methods. Object detection is a fundamental task in computer vision, aiming to identify objects in images. It has evolved significantly, with early methods relying on handcrafted features and later methods leveraging deep learning. The paper reviews milestone detectors, datasets, metrics, and technical advancements, including the transition from traditional methods to deep learning-based approaches. It highlights the development of two-stage and one-stage detectors, the role of feature pyramids, and the integration of attention mechanisms. The paper also discusses challenges such as object localization, speed, and accuracy, and presents recent advancements like DETR and CenterNet. It covers object detection datasets, evaluation metrics, and technical evolution, including multi-scale detection, context priming, hard negative mining, and loss functions. The paper also addresses speed-up techniques, including feature map sharing, cascaded detection, network pruning, lightweight network design, and numerical acceleration. Overall, the survey provides a detailed overview of the evolution of object detection, emphasizing its technical progress and future directions.