Comprehensive Survey on Deep Learning-Based Advances in One-Stage Object Detection

Object detection is a fundamental problem in computer vision, with applications spanning self-driving cars, surveillance systems, medical imaging, robotics, and smart cities. Among the myriad of algorithms developed for this task, the You Only Look Once (YOLO) family stands out for its ability to perform real-time and accurate object detection. This article provides a comprehensive analysis of the YOLO algorithm series, from YOLOv1 to YOLOv8, evaluating them across key performance metrics, including precision, recall, mean Average Precision (mAP), frames per second (FPS), and overall effectiveness. Unlike traditional two-stage detectors such as R-CNN, YOLO formulates object detection as a single regression problem: a single pass over the image simultaneously predicts bounding boxes and class probabilities. This end-to-end design enables YOLO to achieve high speed while maintaining a competitive accuracy-efficiency trade-off. We examine architectural innovations across YOLO versions, including batch normalization, anchor boxes, residual blocks, feature pyramid networks, and attention mechanisms, and discuss their impact on performance. Lightweight models (e.g., YOLOv5-Nano, YOLOv8-Small) are explored with a focus on their suitability for mobile and embedded systems, highlighting YOLO’s adaptability to resource-constrained environments. Challenges such as small object detection, occlusion, and domain-specific tuning are also addressed. This article serves as a practical guide for researchers, developers, and practitioners aiming to leverage YOLO for real-world object detection tasks.

Sea turtles exhibit high migratory rates and occupy a broad range of habitats, which in turn makes monitoring these taxa challenging. Applying deep learning (DL) models to vast image datasets collected from citizen science programs can offer promising solutions to overcome the challenge of monitoring the wide habitats of wildlife, particularly sea turtles. Among DL models, object detection models, such as the You Only Look Once (YOLO) series, have been extensively employed for wildlife classification. Despite their successful application in this domain, detecting objects in images with complex backgrounds, including underwater environments, remains a significant challenge. Recently, instance segmentation models have been developed to address this issue by providing more accurate classification of complex images compared to traditional object detection models. This study compared the performance of two state-of-the-art DL methods namely; the object detection model (YOLOv5) and instance segmentation model (YOLOv5-seg), to detect and classify sea turtles. The images were collected from iNaturalist and Google and then divided into 64% for training, 16% for validation, and 20% for test sets. Model performance during and after finishing training was evaluated by loss functions and various indexes, respectively. Based on loss functions, YOLOv5-seg demonstrated a lower error rate in detecting rather than classifying sea turtles than the YOLOv5. According to mean Average Precision (mAP) values, which reflect precision and recall, the YOLOv5-seg model showed superior performance than YOLOv5. The mAP0.5 and mAP0.5:0.95 for the YOLOv5 model were 0.885 and 0.795, respectively, whereas for the YOLOv5-seg, these values were 0.918 and 0.831, respectively. In particular, based on the loss functions and classification results, the YOLOv5-seg showed improved performance for detecting rather than classifying sea turtles compared to the YOLOv5. The results of this study may help improve sea turtle monitoring in the future.

In this work, we present our intelligent SSD, SeeSSD , an energy-efficient computational SSD for a real-time object detection system. SeeSSD embeds an FPGA-based CNN processing engine and the firmware that performs the convolutional operation on the target image. SeeSSD processes the image data at the storage before sending it to the host. This reduces the amount of data transferred to the host and lowers the data movement overhead, thus reducing transfer time and saving power. By using our SeeSSD system and YOLO_Embed, an object detection neural network model, we are able to outperform the fastest YOLO model for an embedded controller, YOLO-Lite, in terms of performance, accuracy, and energy efficiency. YOLO (You Only Look Once) models are a series of one-stage object detection neural models that have become very popular due to their fast speed and high accuracy. The contribution of this work includes designing and implementing our SeeSSD system with a lightweight object detection model, YOLO_Embed, for reducing the data movement overhead, performing real-time inference, and lowering the overall power consumption. We implemented the entire software stack associated with the SeeSSD system; on-device CNN acceleration engine implemented on FPGA, object identification interface for SeeSSD using YOLO_Embed, and embedded software layer in SeeSSD for on-device convolutional processing. We calculated our YOLO_Embed model’s accuracy on object detection dataset benchmarks such as PASCAL VOC 2012, which came out to be 38.1% mAP (mean Accuracy Precision). Our system was able to perform inference in 0.21 seconds while reducing the power consumption by approximately 1.2 × and 1.4 × for CPU-Only and CPU+GPU systems, respectively. We were also able to reduce the data movement overhead by 24 × for a single target image.

In the "blind review" mode for technical bid evaluation within the power industry, review experts must make impartial judgments based on anonymized technical response documents. However, these documents often contain a large amount of complex graphical content, which existing general Optical Character Recognition (OCR) technology struggles to accurately identify. To address this, this study first constructs a specialized image dataset covering typical power industry response document scenarios. Subsequently, an improved You Only Look Once (YOLO) object detection model is developed, incorporating an anchor box clustering strategy optimized for small, densely packed power symbols and a multi-scale feature fusion mechanism. Then, a local contextual relationship inference module based on Graph Neural Networks is designed and integrated. Finally, an end-to-end recognition pipeline from image input to structured information output is constructed. Experimental results indicate that the improved model achieves a mean Average Precision (mAP) of 90.0% in key equipment detection tasks; in topology relationship restoration experiments, the average Graph Edit Distance (GED) is 0.172; in end-to-end efficiency experiments simulating real blind review scenarios, the average processing time per document is only 8.7 seconds, with an information completeness rate of 98.2%. The research demonstrates that the YOLO-based object detection and relationship modeling method, customized for the blind review mode, can effectively address the challenge of recognizing complex image content in power response documents, providing reliable technical support for the efficiency and fairness of blind reviews.

The chapter discusses the design and development of a novel object detection model keeping the motive that anyone and everyone with little to no prior knowledge can easily use it. The object detection is a part of computer vision which deals with detection and localization of objects in images or videos. These systems can be implemented for surveillance, keeping track of goods, robotics, medical field, self-driving cars, and much more. A lot of organizations are using it for various purposes such as it is used by NASA for analysis for deep space analysis, self-driving cars use it for making accurate decisions, and airports use it for security and surveillance. Although it has numerous use cases, but because of complex implementation and need of expertise in the field, it is not being widely used currently. In this chapter, we have made an effort to automate most of the process that a user should do in order to train and use models for object detection. The most complex part of using these technologies is training custom models, and only people having expertise in field are doing it. With the system designed by us, anyone would be able to train custom models on their own and they don’t have to do much for it. Normally, first users have to gather large amount of images and then annotate them 124one by one, which is a tedious and time taking process. One has to do this for thousands of images and each of those images can have multiple instances of various objects to train a model to give a good accuracy. We are automating the whole process with the help of AWS recognition module of AWS. You only look once (YOLO) object detection model is used for automation, as it is fast and accurate at the same time which gives it edge over the other models available for use. This model is based on deep learning to learn through the images provided by user in training phase and then later on can be used to perform detection on images or real-time videos. A graphic user interface (GUI) is also developed, so the users can easily interact with the system and train their own custom models and use as they want it to be used. Data are sent over the AWS servers and retrieve the results using a block of python code, then the response collected from AWS is manipulated and the data are extracted that is required to annotate the images and then processing is done according to the need of our target model. This format can be different for various models which can be used for object detection.

Object detection is challenging in the computer vision area and is crucial in autonomous driving systems. The largest number of traditional techniques or the use of deep learning are evaluated in the urban area, but in rural areas, there is little research carried out. The goal of this work is object detection in rural roads, this paper presents the use of deep learning frameworks used as You Only Look Once (YOLO) and another that belongs to the same category of one-stage is the well-known Single Shot Multi-Box Detector (SSD), in the state of the literature, produces excellent results in detecting objects in real-time. These models behave differently in network architecture, for this reason, we use images of rural roads with different environments to achieve an optimal balance between precision and precision in the detection of objects. Furthermore, as there is no dataset in these environments, we created our own data set to perform the experiments due to the difficulty of this problem. The result of both detectors has produced acceptable results under certain conditions like lighting conditions, viewing perspectives, partial occlusion of the object.

Integrating autonomous haulage systems into underground mining has revolutionized safety and operational efficiency. However, deploying 3D detection systems for autonomous truck navigation in such an environment faces persistent challenges due to dust, occlusion, complex terrains, and low visibility. This affects their reliability and real-time processing. While existing reviews have discussed object detection techniques and sensor-based systems, providing valuable insights into their applications, only a few have addressed the unique underground challenges that affect 3D detection models. This review synthesizes the current advancements in 3D object detection models for underground autonomous truck navigation. It assesses deep learning algorithms, fusion techniques, multi-modal sensor suites, and limited datasets in an underground detection system. This study uses systematic database searches with selection criteria for relevance to underground perception. The findings of this work show that the mid-level fusion method for combining different sensor suites enhances robust detection. Though YOLO (You Only Look Once)-based detection models provide superior real-time performance, challenges persist in small object detection, computational trade-offs, and data scarcity. This paper concludes by identifying research gaps and proposing future directions for a more scalable and resilient underground perception system. The main novelty is its review of underground 3D detection systems in autonomous trucks.

Visual tracking is a fundamental technology in computer vision, with wide-ranging applications in fields such as robotics, autonomous driving, augmented reality, and security systems. A key challenge in these domains is ensuring stable object tracking in dynamic and complex environments. Tracking objects in real-world settings presents difficulties like occlusion, lighting changes, and sudden object movements, all of which demand the development of robust algorithms that can handle such variability effectively. This research introduces a novel single object visual tracking technique that leverages the YOLO (You Only Look Once) object detection model, combined with an adaptive particle filter. By integrating YOLO's precise object detection capabilities with the continuous state estimation offered by particle filters, this approach achieves stable object tracking, even in challenging environments. At the heart of the proposed algorithm is a dynamic fusion of YOLO detection results with image feature-based particle weight updates. When YOLO successfully detects an object, the position and size data of the object are used to adjust the state and weights of the particles, resulting in improved tracking accuracy and quick recovery from events such as sudden movements or occlusions. Conversely, in situations where YOLO detection is unavailable, the algorithm seamlessly transitions to a pure particle filter mode. In this fallback mode, visual features, such as image color histograms, are used to update particle weights, ensuring continuous and robust tracking even during temporary detection failures. The object’s dynamic state is modeled using an 8-dimensional vector, which includes information about position, velocity, size, and size-change rate. This comprehensive state representation enables precise tracking of various object motions and size transformations. Additionally, a Bhattacharyya distance-based weight calculation method is employed to assess particle similarity more effectively. Experimental results demonstrate that the proposed method excels in various challenging scenarios, including those with sudden object movements, partial occlusions, and lighting variations. The algorithm consistently delivers stable tracking, underscoring its robustness and adaptability. The contributions of this research have potential applications in fields such as autonomous driving, SLAM (Simultaneous Localization and Mapping), and augmented reality. Specifically, this technique could enhance pedestrian and vehicle tracking in autonomous driving systems and improve the accuracy of dynamic object tracking and environmental mapping in SLAM. It is also likely to play a key role in real-world applications like person tracking in security systems, player motion analysis in sports, and focused tracking in medical imaging. Future research directions include expanding the method to handle multi-object tracking and optimizing its real-time performance, broadening its applicability. Additionally, further improvements in tracking accuracy and robustness may be achieved by incorporating deep learning-based feature extraction and integrating data from various sensors.

One popular of object detection model for object detection is You Only Look Once (YOLO) with humans are among the most often utilized for detection objects. Despite the various of human datasets, just a few research that compared the datasets performance against various versions of the YOLO algorithm. This study compares the performance of YOLOv10, YOLOv11, and YOLOv12 on eight different datasets, such as CrowdHuman, CityPersons, Wider Person, Mall Dataset, INRIA, Microsoft Common Object (MS COCO), PASCAL VOC, and MOT17. Precision, recall, mAP@50, and mAP@50-95 are used to measure the YOLO model version's performance on each dataset. The results indicate that each datasets have different perfomance on each version of YOLO, so the performance on model depends on the variation of the dataset. The best results on the MOT17 dataset are obtained by YOLOv12, with 0.909 in precision, 0.775 in recall, 0.88 in mAP@50, and 0.695 in mAP@50-95. On the City Person dataset. However, YOLOv11 performs best result, with 0.782 in precision, 0.529 in recall, 0.694 in mAP@50, and 0.476 in mAP@50-95. Therefore, choosing a YOLO version that is appropriate for the dataset's complexity is essential to creating the best detection model Therefore, selecting the appropriate YOLO version according to the dataset complexity is crucial to obtain the most optimal detection model.

We propose a lightweight detection algorithm based on the Single Shot MultiBox Detector (SSD) algorithm in order to facilitate sheep management and to realize sheep facial identification, and we take the self-constructed dataset as the research object. First, the SSD replaces the VGG16 backbone network with MobileNetv3, a lightweight neural network, to create a hybrid model that is much smaller. Second, the ECA attention mechanism is incorporated into the backend of the 72 × 160 bottleneck layer. Finally, the SmoothL1 loss function is substituted with the BalancedL1 loss function. The optimized model’s size decreases significantly from the original SSD’s 132 MB to just 22.4 MB. It achieves a mean average precision of 83.47% and maintains an average frame rate of 68.53 frames per second. Compared to the basic SSD model, the mean average precision has increased by 3.25 percentage points, the model size has decreased by 109.6 MB, and the detection speed has improved by 9.55 frames per second. In comparative experiments using the same dataset with different object detection models, the proposed model outperforms the SSD, Faster R-CNN, Retinanet, and CenterNet in terms of mean average precision, with improvements of 3.25 percentage points, 4.71 percentage points, 2.38 percentage points, and 8.13 percentage points, respectively. The detection speed has shown significant improvements, increasing by 9.55, 58.55, 53.1, and 12.37 frames per second, respectively. The improved model presented in this paper significantly reduces the model’s size and computational requirements while maintaining an excellent performance. This provides a valuable reference for the digitalization of animal husbandry and livestock farming.

Wild birds are monitored with the important objectives of identifying their habitats and estimating the size of their populations. Especially in the case of migratory bird, they are significantly recorded during specific periods of time to forecast any possible spread of animal disease such as avian influenza. This study led to the construction of deep-learning-based object-detection models with the aid of aerial photographs collected by an unmanned aerial vehicle (UAV). The dataset containing the aerial photographs includes diverse images of birds in various bird habitats and in the vicinity of lakes and on farmland. In addition, aerial images of bird decoys are captured to achieve various bird patterns and more accurate bird information. Bird detection models such as Faster Region-based Convolutional Neural Network (R-CNN), Region-based Fully Convolutional Network (R-FCN), Single Shot MultiBox Detector (SSD), Retinanet, and You Only Look Once (YOLO) were created and the performance of all models was estimated by comparing their computing speed and average precision. The test results show Faster R-CNN to be the most accurate and YOLO to be the fastest among the models. The combined results demonstrate that the use of deep-learning-based detection methods in combination with UAV aerial imagery is fairly suitable for bird detection in various environments.

Small object detection is a challenging computer vision problem due to their low feature representation in the images and factors such as occlusions and noise. In images captured from a camera mounted on an unmanned aerial vehicle (UAV), objects are usually acquired in small sizes depending on the UAV flight altitude. The state-of-the-art object detectors often have lower detection accuracy with small objects. New approaches of combining features at multi-levels in the network helps in improving the object detection performance. In this paper, we propose a multi-scale approach of low-level feature combinations with deconvolutional modules on a single shot multibox detection (SSD) object detector to improve the small object detection in images acquired from a UAV. The proposed SSD based architecture is evaluated on UAV datasets to compare its performance with the state-of-the-art detectors.

Object detection is a challenging task in remote sensing because objects only occupy a few pixels in the images, and the models are required to simultaneously learn object locations and detection. Even though the established approaches well perform for the objects of regular sizes, they achieve weak performance when analyzing small ones or getting stuck in the local minima (e.g., false object parts). Two possible issues stand in their way. First, the existing methods struggle to perform stably on the detection of small objects because of the complicated background. Second, most of the standard methods used handcrafted features and do not work well on the detection of objects parts that are missing. We here address the above issues and propose a new architecture with a multipatch feature pyramid network (MPFP-Net). Different from the current models that, during training, only pursue the most discriminative patches, in MPFP-Net, the patches are divided into class-affiliated subsets, in which the patches are related, and based on the primary loss function, a sequence of smooth loss functions is determined for the subsets to improve the model for collecting small object parts. To enhance the feature representation for patch selection, we introduce an effective method to regularize the residual values and make the fusion transition layers strictly norm-preserving. The network contains bottom-up and crosswise connections to fuse the features of different scales to achieve better accuracy compared to several state-of-the-art object detection models. Also, the developed architecture is more efficient than the baselines.

Object detection with deep learning model has achieved good results in many fields, but in some fields that think highly of data privacy, such as medical care, its applications is greatly limited by data. And Federated Learning allows clients to train a model together, while leaving their data in the local, without sharing with the server or other clients. Using the methods of Federated Learning, such as Federated Averaging(FedAvg), to train models can provide privacy, security benefits. Nonetheless, there is little experiment applying Federated Learning algorithms to train the model with a large number of parameters, such as deep learning object detection model. With non-IID data, the accuracy of object detection model trained by FedAvg reduces significantly, and need more rounds to coverage. In this work, we use Kullback-Leibler divergence(KLD) measure the weights divergence between different model trained with non-IID data. And we propose a useful scheme to improve FedAvg based Abnormal Weights Supression, reducing the influence of the weights divergence caused by non-IID and unbalanced data. As a representative of object detection, we choose Single Shot MultiBox Detector(SSD) as the base model. The results of the experiments show that the Mean Average Precision(mAP) get obvious improvement in Pascal VOC 2007 test dataset.

Object detection is a critical component in intelligent surveillance, disaster response, and autonomous dronebased monitoring systems. In recent years, the You Only Look Once (YOLO) family of algorithms has transformed real-time object detection by achieving an effective balance between detection speed and accuracy. This review paper presents a comprehensive analysis of YOLO models from YOLOv1 to YOLOv8, emphasizing their architectural evolution, backbone development, and feature extraction strategies. Each version’s key innovations are discussed in detail—from YOLOv1’s unified detection framework and YOLOv2’s introduction of anchor boxes to YOLOv8’s advanced anchor-free detection mechanism. The comparative evaluation focuses on model accuracy (mAP), inference speed (FPS), and deployment feasibility for embedded and UAV-based systems. Findings from existing literature and experimental insights reveal that successive YOLO versions have progressively enhanced small object detection, robustness to scale variation, and computational efficiency. Overall, the study concludes that YOLOv8 currently achieves the most optimal trade-off between precision, speed, and resource utilization, establishing it as a leading framework for real-time aerial human detection and a promising foundation for future advancements in autonomous surveillance and rescue applications.