On the Performance of One-Stage and Two-Stage Object Detectors in Autonomous Vehicles Using Camera Data
Open accessThe DOI matches its Crossref record; 46 of the 46 references it cites were found in a registry or an index.
Abstract
Object detection using remote sensing data is a key task of the perception systems of self-driving vehicles. While many generic deep learning architectures have been proposed for this problem, there is little guidance on their suitability when using them in a particular scenario such as autonomous driving. In this work, we aim to assess the performance of existing 2D detection systems on a multi-class problem (vehicles, pedestrians, and cyclists) with images obtained from the on-board camera sensors of a car. We evaluate several one-stage (RetinaNet, FCOS, and YOLOv3) and two-stage (Faster R-CNN) deep learning meta-architectures under different image resolutions and feature extractors (ResNet, ResNeXt, Res2Net, DarkNet, and MobileNet). These models are trained using transfer learning and compared in terms of both precision and efficiency, with special attention to the real-time requirements of this context. For the experimental study, we use the Waymo Open Dataset, which is the largest existing benchmark. Despite the rising popularity of one-stage detectors, our findings show that two-stage detectors still provide the most robust performance. Faster R-CNN models outperform one-stage detectors in accuracy, being also more reliable in the detection of minority classes. Faster R-CNN Res2Net-101 achieves the best speed/accuracy tradeoff but needs lower resolution images to reach real-time speed. Furthermore, the anchor-free FCOS detector is a slightly faster alternative to RetinaNet, with similar precision and lower memory usage.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Figures






Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 2×
“The Deep Residual Networks (ResNet) [20] succeeded in addressing complex image problems achieving the best results in competitions such as ILSVRC [39] and COCO [13].”
BackgroundMethodVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 1×
“Faster R-CNN uses a convolutional backbone network, such as VGG [19] or ResNet [20], which outputs global feature maps.”
MethodLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 2×
“ResNet, ResNeXt, and Res2Net The Deep Residual Networks (ResNet) [20] succeeded in addressing complex image problems achieving the best results in competitions such as ILSVRC [39] and COCO [13].”
BackgroundRich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
(2016) · Cited by 31,590
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939 · cited 1×
“For instance, feature pyramid networks (FPN) were proposed in order to crop RoI features from different levels depending on the scale [21].”
BackgroundMobileNetV2: Inverted Residuals and Linear Bottlenecks
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00474 · Cited by 28,755
Cited by
Ali, M. L., & Zhang, Z. (2024) · Computers · doi:10.3390/computers13120336 · Cited by 383
Sharma, A., Kumar, V., & Longchamps, L. (2024) · Smart Agricultural Technology · doi:10.1016/j.atech.2024.100648 · Cited by 212
Comparing YOLOv8 and Mask R-CNN for instance segmentation in complex orchard environments
Sapkota, R., Ahmed, D., & Karkee, M. (2024) · Artificial Intelligence in Agriculture · doi:10.1016/j.aiia.2024.07.001 · Cited by 200
“…estimation (Lv et al., 2022) 2022 YOLO-V5 Identi fi cation method for the apple growth pattern in the orchard (Su et al., 2022) 2022 YOLO-V5 Tree trunk and obstacle detection in apple orchards (Jia et al., 2022) 2022 Mask R-CNN Ripe and green apple segmentation in orchards (Cong et al., 2022)…”
Ali, R., Kang, D., Suh, G., & Cha, Y.-J. (2021) · Automation in Construction · doi:10.1016/j.autcon.2021.103831 · Cited by 166
Deep learning for visual recognition and detection of aquatic animals: A review
Li, J., Xu, W., Deng, L., Xiao, Y., Han, Z., & Zheng, H. (2022) · Reviews in Aquaculture · doi:10.1111/raq.12726 · Cited by 150
Dananjayan, S., Tang, Y., Zhuang, J., Hou, C., & Luo, S. (2022) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106658 · Cited by 158
A novel deep learning‐based method for detection of weeds in vegetables
Jin, X., Sun, Y., Che, J., Bagavathiannan, M., Yu, J., & Chen, Y. (2022) · Pest Management Science · doi:10.1002/ps.6804 · Cited by 136
“70 Consequently, the low detection speed of the two-stage detector Faster R-CNN model may limit its potential applications.”
BackgroundDemetriou, D., Mavromatidis, P., Robert, P. M., Papadopoulos, H., Petrou, M., & Nicolaides, D. (2023) · Waste Management · doi:10.1016/j.wasman.2023.05.039 · Cited by 137
Mao, M., & Hong, M. (2025) · Sensors · doi:10.3390/s25072270 · Cited by 161
“On the other hand, two-stage detectors, such as Faster R-CNN [47], generate regional proposals and refine these proposals in a second stage for classification and localization [48].”
Tan, M., Chao, W., Cheng, J.-K., Zhou, M., Ma, Y., Jiang, X., Ge, J., Yu, L. Z., & Feng, L. (2022) · Animals · doi:10.3390/ani12151976 · Cited by 135
“It has an anchor-based one-stage detector with a fast inference speed [38].”
BackgroundVehicle Detection from Aerial Images Using Deep Learning: A Comparative Study
Ammar, A., Koubâa, A., Ahmed, M., Saad, A., & Benjdira, B. (2021) · Electronics · doi:10.3390/electronics10070820 · Cited by 116
“We selected in the first branch, Faster R-CNN [19], which is the most representative model from the two-stage family, according to [34].”
MethodRendón-Segador, F. J., Álvarez-García, J. A., Enríquez, F., & Déniz, Ó. (2021) · Electronics · doi:10.3390/electronics10131601 · Cited by 86
Deep transfer learning for intelligent vehicle perception: A survey
Liu, X., Li, J., Ma, J., Sun, H., Xu, Z., Zhang, T., & Yu, H. (2023) · Green Energy and Intelligent Transportation · doi:10.1016/j.geits.2023.100125 · Cited by 85
Enhanced pothole detection system using YOLOX algorithm
Mohan Prakash, B., & Sriharipriya, K. C. (2022) · Autonomous Intelligent Systems · doi:10.1007/s43684-022-00037-z · Cited by 83
“Object detection algorithms using deep learning can be classified between one-stage detectors and two-stage detectors [44].”
BackgroundA comprehensive review of object detection with traditional and deep learning methods
Pagire, V. R., Chavali, M. S., & Kale, A. (2025) · Signal Processing · doi:10.1016/j.sigpro.2025.110075 · Cited by 79
Object-level benchmark for deep learning-based detection and classification of weed species
Hasan, A. S. M. M., Diepeveen, D., Laga, H., Jones, M. G. K., & Sohel, F. A. (2023) · Crop Protection · doi:10.1016/j.cropro.2023.106561 · Cited by 71
Jia, W., Lu, Z., Zhang, H., Liu, Z., Wang, J., & Qu, G. (2022) · Network and Distributed System Security Symposium · doi:10.14722/ndss.2022.24130 · Cited by 99
Moshayedi, A. J., Roy, A. S., Taravet, A., Liao, L., Wu, J., & Gheisari, M. (2023) · Future Transportation · doi:10.3390/futuretransp3010012 · Cited by 55
Evaluation Method of Deep Learning-Based Embedded Systems for Traffic Sign Detection
Lopez-Montiel, M. A., Orozco-Rosas, U., Sánchez-Adame, M., Picos, K., & Montiel, O. (2021) · IEEE Access · doi:10.1109/access.2021.3097969 · Cited by 61
“The two-stage methods divide the detection process into the region proposal and the classification stage [11].”
MethodZarei, N., Moallem, P., & Shams, M. (2022) · IEEE Access · doi:10.1109/access.2022.3221942 · Cited by 60
“…revolutionized the detection The associate editor coordinating the review of this manuscript and approving it for publication was Angel F. García-Fernández . of objects; however, despite recent advances, the tradeoff between the speed and accuracy of such networks remains a major challenge [6].”
BackgroundCarranza-García, M., Lara-Benítez, P., García–Gutiérrez, J., & Riquelme, J. C. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.04.001 · Cited by 53
Self-citationA deep learning approach for insulator instance segmentation and defect detection
Antwi-Bekoe, E., Liu, G., Ainam, J.-P., Sun, G., & Xie, X. (2022) · Neural Computing and Applications · doi:10.1007/s00521-021-06792-z · Cited by 50
“Our choice of two-stage detection or mask prediction model over single-stage model is influenced by the fact that two-stage detection models are much more scalable, proven to offer more competitive accuracy for higher resolution input images in most architectures, much more robust to the presence of imbalanced data as…”
BackgroundAnalysis of Automotive Camera Sensor Noise Factors and Impact on Object Detection
Li, B., Chan, P. H., Baris, G., Higgins, M. D., & Donzella, V. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3211406 · Cited by 51
“Furthermore, camera data processing can rely on well-established and mature computer vision and machine learning techniques [11].”
MethodEdge Computing-Driven Real-Time Drone Detection Using YOLOv9 and NVIDIA Jetson Nano
Hakani, R., & Rawat, A. (2024) · Drones · doi:10.3390/drones8110680 · Cited by 67
Real-time tool detection in smart manufacturing using You-Only-Look-Once (YOLO)v5
Zendehdel, N., Chen, H., & Leu, M. C. (2023) · Manufacturing Letters · doi:10.1016/j.mfglet.2023.08.062 · Cited by 40
Neto, A., Camara, J., & Cunha, A. M. T. D. S. (2022) · Sensors · doi:10.3390/s22041449 · Cited by 38
“This deep residual network exploits residual blocks to overcome gradient gradation [23,26].”
BackgroundWang, N., Luo, Y., Sato, T., Xu, K., & Chen, Q. A. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00407 · Cited by 81
“Camera-based AD perception generally leverages DNN-based object detection to detect or recognize road objects of various categories (e.g., traffic signs, vehicles, and pedestrians) in consecutive image frames [10].”
BackgroundJiang, C., Wang, Z., Liang, H., & Tan, S. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3155232 · Cited by 38
“In the research on object detection of mobile robots, real-time detection is always a technical difficulty that limits the application of algorithms [23].”
BackgroundA survey on 3D object detection in real time for autonomous driving
Contreras, M., Jain, A., Bhatt, N. P., Banerjee, A., & Hashemi, E. (2024) · Frontiers in Robotics and AI · doi:10.3389/frobt.2024.1212070 · Cited by 43
Karbouj, B., Topalian-Rivas, G. A., & Krüger, J. (2024) · Procedia CIRP · doi:10.1016/j.procir.2024.01.077 · Cited by 38
Object detection using depth completion and camera-LiDAR fusion for autonomous driving
Carranza-García, M., Galán-Sales, F. J., Luna-Romera, J. M., & Riquelme, J. C. (2022) · Integrated Computer-Aided Engineering · doi:10.3233/ica-220681 · Cited by 37
“As found in previous studies with this dataset [31], the two-stage Faster R-CNN architecture provides the best detection accuracy.”
InfluentialSelf-citationLiCaNet: Further Enhancement of Joint Perception and Motion Prediction Based on Multi-Modal Fusion
Khalil, Y. H., & Mouftah, H. T. (2022) · IEEE Open Journal of Intelligent Transportation Systems · doi:10.1109/ojits.2022.3160888 · Cited by 34
Moving vehicle detection and tracking at roundabouts using deep learning with trajectory union
Avşar, E., & ÖZİNAL, Y. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-021-11804-0 · Cited by 31
“Both of these methods have been developed at recent times have useful applications like moving object detection [25] and object detection in autonomous vehicles [11].”
MethodLong-Distance Vehicle Dynamic Detection and Positioning Based on Gm-APD Lidar and LIDAR-YOLO
Ding, Y., Qu, Y., Du, D., Jiang, Y., Zhang, H., Song, B., Zhou, X., & Sun, J. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3193740 · Cited by 28
Detecting common coccinellids found in sorghum using deep learning models
Wang, C., Grijalva, I., Caragea, D., & McCornack, B. P. (2023) · Scientific Reports · doi:10.1038/s41598-023-36738-5 · Cited by 29
“Traditionally, two-stage detectors have been more accurate than one-stage detectors, while the one-stage detectors have been faster and more suitable for use in practical applications that require real-time object detection 32 .”
Shi, P., Jiang, Q., Shi, C., Xi, J., Tao, G., Zhang, S., Zhang, Z., Liu, B., Gao, X., & Wu, Q. (2021) · Remote Sensing · doi:10.3390/rs13163243 · Cited by 25
Comparison of YOLOv5 and YOLOv6 Models for Plant Leaf Disease Detection
İren, E. (2024) · Engineering Technology & Applied Science Research · doi:10.48084/etasr.7033 · Cited by 24
“Non-max suppression is a common algorithm used for discarding bounding boxes when multiple boxes are predicted for the same object [9-10].”
MethodAulia, U., Hasanuddin, I., Dirhamsyah, M., & Nasaruddin, N. (2024) · Heliyon · doi:10.1016/j.heliyon.2024.e35247 · Cited by 38
“As a result, the elapsed time for the training process for the proposed model was lower than for comparable models in other studies [53].”
ResultRyu, S.-E., & Chung, K. (2021) · Applied Sciences · doi:10.3390/app11157093 · Cited by 26
“Although such models are suitable for real-time detection, they are limited i that their accuracy is relatively low[14].”
BackgroundInfluentialVerschoof, W. B., Bonhage, A., Schneider, A., Ouimet, W. B., & Raab, T. (2022) · Archaeological Prospection · doi:10.1002/arp.1889 · Cited by 27
Research on Individual Recognition and Matching of Whale and Dolphin Based on EfficientNet Model
Wu, S., Wang, J., Ping, Y., & Zhang, X. (2022) · International Conference on Big Data, Artificial Intelligence and Internet of Things Engineering (ICBAIE) · doi:10.1109/icbaie56435.2022.9985881 · Cited by 21
“In order to improve efficiency, one-stage methods have been gradually developed, such as Yolo, SqueezeDet and DetectNet models [11].”
MethodImproving accuracy and efficiency in seagrass detection using state-of-the-art AI techniques
Noman, M. K., Islam, S. M. S., Abu-Khalaf, J., Jalali, S. M. J., & Lavery, P. S. (2023) · Ecological Informatics · doi:10.1016/j.ecoinf.2023.102047 · Cited by 20
Real-time safety detection on construction sites using a vision-language and NLP-based model
Sivanraj, S., Uduwage, D. N. L. S., & Tripathi, M. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2025.103889 · Cited by 30
Habash, N. M., Alqumsan, A. A., & Zhou, T. (2025) · Sensors · doi:10.3390/s25247563 · Cited by 35
Computer-vision based analysis of the neurosurgical scene – A systematic review
Buyck, F., Vandemeulebroucke, J., Ceranka, J., Gestel, F. V., Cornelius, J. F., Duerinck, J., & Bruneau, M. (2023) · Brain and Spine · doi:10.1016/j.bas.2023.102706 · Cited by 19
Recent Advances in 3D Object Detection for Self-Driving Vehicles: A Survey
Fawole, O. A., & Rawat, D. B. (2024) · AI · doi:10.3390/ai5030061 · Cited by 31
“Although effective in recognizing objects on the visual plane, traditional 2D detection techniques [22] do not provide the depth information required for driving tasks such as path planning and collision avoidance.”
BackgroundOcholla, I. A., Pellikka, P., Karanja, F., Vuorinne, I., Väisänen, T., Boitt, M. K., & Heiskanen, J. (2024) · Remote Sensing · doi:10.3390/rs16162929 · Cited by 20
HookNet: Efficient Multiscale Context Aggregation for High-Accuracy Detection of Fabric Defects
Xiang, Z., Shen, Y., Ma, M., & Qian, M. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3280519 · Cited by 20
Enhanced SSD Algorithm-Based Object Detection and Depth Estimation for Autonomous Vehicle Navigation
Saini, V., Kantipudi, M. P., & Meduri, P. (2023) · International Journal of Transport Development and Integration · doi:10.18280/ijtdi.070408 · Cited by 18
“Carranza-García et al. [19] contrasted single-stage like Yolo V3 and two-stage detectors like Faster R-CNN.”
BackgroundA compact deep architecture for real-time saliency prediction
Zabihi, S., Tavakoli, H. R., Borji, A., & Mansoori, E. G. (2022) · Signal Processing Image Communication · doi:10.1016/j.image.2022.116671 · Cited by 17
A Survey on Real-Time Object Detection on FPGAs
Hozhabr, S. H., & Giorgi, R. (2025) · IEEE Access · doi:10.1109/access.2025.3544515 · Cited by 24
“As the name implies, these detectors are made up of two major stages : 1) Region proposal: to propose a set of candidate regions or bounding boxes likely to contain objects of interest [111].”
Enhanced Multi-Task Learning Architecture for Detecting Pedestrian at Far Distance
Zhou, C., Wu, M., & Lam, S.-K. (2022) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2022.3142445 · Cited by 15
“As reported in [38], if the pedestrian detection model fails to achieve real-time processing (i.e., 10 frames per second (FPS) on Waymo dataset [39]), some important frames will be discarded, which can severely impact the safety decisions made by the autonomous systems.”
BackgroundKumar, P., & Misra, U. (2024) · International Conference on Disruptive Technologies (ICDT) · doi:10.1109/icdt61202.2024.10489628 · Cited by 16
A Comprehensive Study on Self-Learning Methods and Implications to Autonomous Driving
Xing, J., Wei, D., Zhou, S., Wang, T., Huang, Y., & Chen, H. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3440498 · Cited by 27
The Impact of YOLO Algorithms Within Fall Detection Application: A Review
Khekan, A. R., Aghdasi, H. S., & Salehpour, P. (2024) · IEEE Access · doi:10.1109/access.2024.3496823 · Cited by 23
Hybrid Reasoning Based on Large Language Models for Autonomous Car Driving
Azarafza, M., Nayyeri, M., Steinmetz, C., Staab, S., & Rettberg, A. (2024) · International Conference on Control, Mechatronics and Automation (ICCMA) · doi:10.1109/iccma63715.2024.10843921 · Cited by 34
“CNN-based object detection approaches are mainly divided into two kinds: single-stage and two-stage object detection methods [23].”
Detecting Plant Diseases Using Machine Learning Models
Kohut, N., Basystiuk, O., Shakhovska, N., & Melnykova, N. (2024) · Sustainability · doi:10.3390/su17010132 · Cited by 22
R-Sparse R-CNN: SAR Ship Detection Based on Background-Aware Sparse Learnable Proposals
Kamirul, K., Pappas, O. A., & Achim, A. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3577766 · Cited by 18
IR Reasoner: Real-time Infrared Object Detection by Visual Reasoning
Gündoğan, M. M., Aksoy, T., Temizel, A., & Halıcı, U. (2023) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw59228.2023.00048 · Cited by 22
“Single Shot MultiBox Detector (SSD) [33] and You Only Look Once (YOLO) [37] are prominent examples [5].”
BackgroundUAV-based automated earthwork progress monitoring using deep learning with image inpainting
Ersöz, A. B., & Pekcan, O. (2025) · Automation in Construction · doi:10.1016/j.autcon.2025.106211 · Cited by 14
An Approximate GEMM Unit for Energy-Efficient Object Detection
Pilipović, R., Risojević, V., Božič, J., Bulić, P., & Lotrič, U. (2021) · Sensors · doi:10.3390/s21124195 · Cited by 16
“Two-stage detectors possess high accuracy rates, but are slow, while single-stage detectors achieve lower accuracy, but work faster [72].”
BackgroundA review on object detection for autonomous mobile robot
Abdul-Khalil, S., Abdul-Rahman, S., Mutalib, S., Kamarudin, S. I., & Kamaruddin, S. S. (2023) · IAES International Journal of Artificial Intelligence · doi:10.11591/ijai.v12.i3.pp1033-1043 · Cited by 17
Baldovino, R. G., Vidad, A. J. P., Abastillas, R. P. B., Bugtai, N. T., Dadios, E. P., Vicerra, R. R. P., Bandala, A. A., See, A. R., & Roxas, N. R. (2024) · Procedia Computer Science · doi:10.1016/j.procs.2024.09.158 · Cited by 13
Nahata, D., & Othman, K. (2023) · AIMS Electronics and Electrical Engineering · doi:10.3934/electreng.2023016 · Cited by 16
Hussain, K., Moreira, C., Pereira, J. M., Jardim, S., & Jorge, J. A. (2025) · Smart Cities · doi:10.3390/smartcities8030079 · Cited by 16
“Additionally, GCNet [72] and two-stage object detection models [84] could be further developed by integrating sensor fusion techniques (e.g., camera and LiDAR data).”
Fast and Robust People Detection in RGB Images
Dumitrescu, F., Boiangiu, C.-A., & Voncilă, M.-L. (2022) · Applied Sciences · doi:10.3390/app12031225 · Cited by 15
Krinitskiy, M., Sprygin, A. A., Elizarov, S., Narizhnaya, A., Shikhov, A. N., & Chernokulsky, А. V. (2023) · Remote Sensing · doi:10.3390/rs15143493 · Cited by 14
Machine vision approach for monitoring and quantifying fish school migration
Lin, F., Zhu, J., You, A., & Hua, L. (2024) · Ecological Indicators · doi:10.1016/j.ecolind.2024.112769 · Cited by 15
Configurable Hardware Core for IoT Object Detection
Miranda, P. R., Pestana, D. G., Lopes, J. D., Duarte, R. P., Véstias, M. P., Neto, H. C., & de Sousa, J. T. (2021) · Future Internet · doi:10.3390/fi13110280 · Cited by 11
“than one-stage detectors but are computationally more demanding [18].”
BackgroundDivyanth, L. G., Khanal, S. R., Paudel, A., Mattupalli, C., & Karkee, M. (2024) · Frontiers in Plant Science · doi:10.3389/fpls.2024.1512632 · Cited by 12
Unlocking the power of AI for phenotyping fruit morphology in Arabidopsis
Atkins, K., Garzón-Martínez, G. A., Lloyd, A. H., Doonan, J. H., & Lü, C. (2024) · GigaScience · doi:10.1093/gigascience/giae123 · Cited by 9
“Two-stage methods, in contrast, have an initial object proposal stage follo w ed b y a classification sta ge, whic h tends to r esult in mor e pr ecise detections [ 22 ] while still gener ating outputs in negligible time.”
Kelly, M., Feirer, S., Hogan, S., Lyons, A., Lin, F., & Jacygrad, E. (2025) · Drones · doi:10.3390/drones9090593 · Cited by 10
“A key aspect of two-shot models is that they divide an image into regions of interest (ROIs) to focus on areas where objects are likely to be present [43].”
Talking Cars, Doubtful Users—A Population Study in Virtual Reality
Derakhshan, S., Nezami, F. N., Wächter, M. A., Czeszumski, A., Keshava, A., Lukanov, H., Palol, M. V. D., Pipa, G., & König, P. (2022) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2022.3168437 · Cited by 11
Wild Animals Detection Based on YOLOv5
Li, X. (2023) · Applied and Computational Engineering · doi:10.54254/2755-2721/8/20230285 · Cited by 9
Overview of deep learning YOLO algorithm
Pan, Y., Wang, G., & Yu, J. (2025) · International Conference on Computer Vision, Application, and Algorithm (CVAA) · doi:10.1117/12.3055712 · Cited by 14
“Its development phases can be broadly classified into two stages: One-Stage and Two-Stage [5] , and the YOLO algorithm falls into the former category.”
Timilsina, S., Jang, S. M., Jo, C. W., Kwon, Y. N., Sohn, K., Lee, K. H., & Kim, J. S. (2023) · Advanced Intelligent Systems · doi:10.1002/aisy.202300314 · Cited by 10
Object Detection to Identify Shapes of Swallow Nests Using a Deep Learning Algorithm
Indrajaya, D., Setiawan, A., Hartanto, D., & Hariyanto, H. (2022) · Khazanah Informatika Jurnal Ilmu Komputer dan Informatika · doi:10.23917/khif.v8i2.16489 · Cited by 12
“FPN itself is an architecture to produce pyramidal features in object detection, whereas FPNLite is a development of FPN which can produce models with lighter detection capabilities when run [19], [20].”
aGRodet 2.0: An Automated Real-Time Approach for Multiclass Plant Disease Detection
Mitra, A., Mohanty, S. P., & Kougianos, E. (2023) · SN Computer Science · doi:10.1007/s42979-023-02076-6 · Cited by 12
Bi-Directional and Triangular Circulation Fusion Neural Networks for Small Object Detection
Li, F., Duan, J., Zhang, Q., Shan, C., & Han, H. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3523923 · Cited by 8
“However, in small object detection, to accurately extract the area where the small object is located is difficult for two-stage detectors [16].”
Deep Learning-Based Lesion Detection in Endoscopy: A Systematic Literature Review
Rochmawati, N., Fatichah, C., Amaliah, B., Raharjo, A. B., Dumont, F., Thibaudeau, E., & Dumas, C. (2025) · IEEE Access · doi:10.1109/access.2025.3548167 · Cited by 12
“There are two main approaches to deep learning-based object detection: the one-stage and the two-stage approaches [77].”
Enhancing UAV object detection with an efficient multi-scale feature fusion framework
Lai, D., Kang, K., Xu, K., Ma, X., Zhang, Y., Huang, F., & Chen, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0332408 · Cited by 11
AWD-YOLO enhancing autonomous driving perception reliability in adverse weather
Yuan, Y., Dong, W., Yang, S., & Wu, T. (2026) · Scientific Reports · doi:10.1038/s41598-025-29575-1 · Cited by 17
Object Detectors in Autonomous Vehicles: Analysis of Deep Learning Techniques
Du, L. (2023) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2023.0141024 · Cited by 9
“These models operate with a single pass over the input data, offering real-time performance [15].”
BackgroundUAV and Deep Learning: Detection of selected riparian species along the Ganga River
Tripathi, R. N., Ramachandran, A., Agarwal, K., Tripathi, V., Badola, R., & Hussain, S. A. (2024) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlviii-1-2024-637-2024 · Cited by 9
“Examples of two-stage detectors include DEtection Transformer (DETR), Faster Region-Based Convolutional Neural Network (R-CNN), Mask R-CNN, Cascade R-CNN, Feature Pyramid Networks (FPN), and Region-based Fully Convolutional Networks (R-FCN) (Carranza-García et al., 2021; Piao et al., 2022).”
BackgroundMethodObject detection algorithms to identify skeletal components in carbonate cores
Dawson, H. L., & John, C. M. (2024) · Marine and Petroleum Geology · doi:10.1016/j.marpetgeo.2024.106965 · Cited by 8
Shamsollahi, D., Moselhi, O., & Khorasani, K. (2024) · Journal of Construction Engineering and Management · doi:10.1061/jcemd4.coeng-15115 · Cited by 8
Automotive DNN-Based Object Detection in the Presence of Lens Obstruction and Video Compression
Baris, G., Li, B., Chan, P. H., Avizzano, C. A., & Donzella, V. (2025) · IEEE Access · doi:10.1109/access.2025.3544773 · Cited by 11
“A popular dual-stage detector, Faster R-CNN, with state-of-the-art performance was selected for this study [53], [54].”
Generative-Discriminative Machine Learning Models for High-Frequency Financial Regime Classification
Koukorinis, A., Peters, G. W., & Germano, G. (2025) · Methodology And Computing In Applied Probability · doi:10.1007/s11009-025-10148-8 · Cited by 9
YOLO-GML: An object edge enhancement detection model for UAV aerial images in complex environments
Zheng, Z., Zhao, J., & Fan, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0328070 · Cited by 9
“Our approach specifically addresses the reduced contrast and blurred boundaries prevalent in foggy conditions by enhancing edge information across multiple scales [25].”
Unifying ground and air: a comprehensive review of deep learning-enabled CAVs and UAVs
Zia, M. U., Xiang, W., Huang, T., Ahmad, J., Chattha, J. N., Naqvi, I. H., & Butt, F. A. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11425-1 · Cited by 13
Utilizing High Resolution Satellite Imagery for Automated Road Infrastructure Safety Assessments
Brkić, I., Ševrović, M., Medak, D., & Miler, M. (2023) · Sensors · doi:10.3390/s23094405 · Cited by 11
Estimating the effect of age-induced defocus on object detection performance of automotive cameras
Pandey, A., Kühn, S., Kettelgerdes, M., Wunderle, B., & Elger, G. (2025) · Results in Engineering · doi:10.1016/j.rineng.2025.105951 · Cited by 11
Fang, F., Xu, Q., Cheng, Y., Sun, Y., & Lim, J. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3171781 · Cited by 8
“This echoes with observations in [16] and [32] that two-stage detector is more robust on images with degraded quality and varying object size.”
BackgroundImproved YOLOv4 approach: A real time occluded vehicle detection
Kumar, S., Jailia, M., & Varshney, S. (2022) · International Journal of Computing and Digital Systems · doi:10.12785/ijcds/120139 · Cited by 7
“One-stage detectors are detection frameworks that recognize objects rapidly using the bounding box’s size, with a dimension of the structuring element purposely designed to be helpful in a variety of contexts [20].”
Olayiwola, J., Akanmu, A. A., Gao, X., Murzi, H., & Afsari, K. (2023) · Journal of Computing in Civil Engineering · doi:10.1061/jccee5.cpeng-5206 · Cited by 10
Smagulova, K., Elsheikh, A., Silva, D., Fouda, M. E., & Eltawil, A. M. (2025) · Frontiers in Robotics and AI · doi:10.3389/frobt.2025.1674421 · Cited by 10
Ilo, B., Badjona, A., Singh, Y., Shenfield, A., & Zhang, H. (2025) · Processes · doi:10.3390/pr13113731 · Cited by 13
Proximity-Based Optical Camera Communication with Multiple Transmitters Using Deep Learning
Nasution, M. R. A., Herfandi, H., Sitanggang, O. S., Nguyen, H. Q., & Jang, Y. M. (2024) · Sensors · doi:10.3390/s24020702 · Cited by 6
“Although the two-stage approach is often considered for its ability to produce more precise detections, it is comparatively slower because of the preprocessing involved in the initial stage [27].”
BackgroundShao, Z., Yang, K., Sun, P., Hu, Y., & Boukerche, A. (2024) · Computer Communications · doi:10.1016/j.comcom.2024.06.015 · Cited by 9
Extraction of Meta-Data for Recommendation Using Keyword Mapping
Kim, G.-W., Kim, W.-H., Chung, K., & Kim, J.-C. (2024) · IEEE Access · doi:10.1109/access.2024.3430375 · Cited by 10
“COCO (Common Objects in Context) [32] is one of the datasets widely used in the field of Object Detection within Computer Vision.”
Method
Metadata
- Authors
- Carranza-García, M., Torres-Mateo, J., Lara-Benítez, P., & García-Gutiérrez, J.
- Affiliations
- Universidad de Sevilla (ES)
- Division of Computer Science, University of Sevilla, ES-41012 Seville, Spain
- Corresponding author
- Carranza-García, Manuel
- Venue
- Remote Sensing
- Publisher
- MDPI AG
- Published
- 2020-12-29
- Volume
- 13
- Issue
- 1
- Pages
- 89
- ISSN
- 2072-4292 (ISSN-L)
- OpenAlex
- W3113769242
- Cited by
- 270
- References
- 40
- Citations by year
- 2026: 34 · 2025: 62 · 2024: 75 · 2023: 42 · 2022: 40 · 2021: 17
- FWCI
- 9.63
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Remote Sensing (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Deposito de Investigacion Universidad de Sevilla (University of Seville) (submittedVersion, cc-by-nc-nd)
- Funders
- Junta de Andalucía (US-126334, P18-RT-2778)
- Ministerio de Economía y Competitividad (TIN2017-88209-C2-2-R)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- object detection, autonomous vehicles, one-stage detectors, two-stage detection, Faster R-CNN, RetinaNet, YOLOv3, FCOS, vehicle detection, pedestrian detection, cyclist detection, feature extractor, speed-accuracy trade-off, Waymo Open Dataset, real-time detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.