You Only Look Once: Unified, Real-Time Object Detection
Open accessThe DOI matches its Crossref record; 53 of the 53 references it cites were found in a registry or an index; 8 of the 25 sources its claims cite were checked for support.
Abstract
We present YOLO, a new approach to object detection. Prior work on object detection repurposes classifiers to perform detection. Instead, we frame object detection as a regression problem to spatially separated bounding boxes and associated class probabilities. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation. Since the whole detection pipeline is a single network, it can be optimized end-to-end directly on detection performance. Our unified architecture is extremely fast. Our base YOLO model processes images in real-time at 45 frames per second. A smaller version of the network, Fast YOLO, processes an astounding 155 frames per second while still achieving double the mAP of other real-time detectors. Compared to state-of-the-art detection systems, YOLO makes more localization errors but is less likely to predict false positives on background. Finally, YOLO learns very general representations of objects. It outperforms other detection methods, including DPM and R-CNN, when generalizing from natural images to other domains like artwork.
Overview
CoChat Summary
TL;DR
YOLO reframes object detection as a single regression problem, using one convolutional network to predict bounding boxes and class probabilities directly from full images. It achieves real-time detection speeds while remaining competitive with, and complementary to, slower region-based detectors.
Key findings
- YOLO runs at 45 FPS on a Titan X GPU, and Fast YOLO runs at 155 FPS; Fast YOLO reaches 52.7% mAP on PASCAL VOC 2007, more than double prior real-time detectors, while YOLO reaches 63.4% mAP.
- On PASCAL VOC 2012, YOLO scores 57.9% mAP; combining Fast R-CNN with YOLO raises Fast R-CNN from 71.8% to 75.0% mAP on VOC 2007, a 3.2% gain.
- YOLO makes fewer background false positives than Fast R-CNN: Fast R-CNN has 13.6% background false positives among top detections and is almost 3× more likely to predict background detections than YOLO.
- YOLO generalizes better to artwork: on Picasso, YOLO gets 53.3 AP versus R-CNN’s 10.4 AP and DPM’s 37.8 AP; on People-Art, YOLO gets 45 AP versus R-CNN’s 26 AP and DPM’s 32 AP.
Methods
- Detection is modeled as a single CNN regression: the image is divided into an S×S grid; for PASCAL VOC, S=7, B=2, C=20, producing a 7×7×30 tensor of bounding boxes, confidence scores, and class probabilities.
- The base network has 24 convolutional layers followed by 2 fully connected layers; Fast YOLO uses 9 convolutional layers. Convolutional layers are pretrained on ImageNet at 224×224, then detection training uses 448×448 input.
- Training uses PASCAL VOC 2007 and 2012 for about 135 epochs, batch size 64, momentum 0.9, decay 0.0005, sum-squared error with λ_coord=5 and λ_noobj=0.5, dropout, and data augmentation.
Limitations
- The paper states YOLO imposes strong spatial constraints because each grid cell predicts only two boxes and one class, limiting nearby objects and struggling with small objects in groups such as flocks of birds.
- The paper states YOLO struggles to generalize to objects in new or unusual aspect ratios or configurations, uses relatively coarse features due to downsampling, and its loss treats small and large box errors equally; its main error source is incorrect localization.
- The paper states YOLO still lags state-of-the-art detection accuracy and struggles to precisely localize some objects, especially small ones.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures





Verification
Claims the paper makes
6 Valid · 8 Not valid · 15 Not checked · 92 Own results and interpretations
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals.”
What the sources say
- Could not verifycites [20] [20] K. Lenc and A. Vedaldi. R-cnn minus r. arXiv preprint arXiv:1506.06981 , 2015. 5, 6 - sourceExperimental — accuracy not yet measured
The ConvNet-based Object Detection section identifies R-CNN minus R as a method that uses predefined regions instead of region proposals. It gives no runtime measurement or assessment of whether the method achieves real-time detection; searches for real-time performance and speed did not establish that claim. (LLM, conf=0.76)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“To avoid overfitting we use dropout and extensive data augmentation.”
What the sources say
- Partly supportedcites [18] [18] G. E. Hinton, N. Srivastava, A. Krizhevsky, I. Sutskever, and R. R. Salakhutdinov. Improving neural networks by preventing co-adaptation of feature detectors. arXiv preprint arXiv:1207.0580 , 2012. 4 - sourceExperimental — accuracy not yet measured
The paper states that dropout reduces overfitting (p. 1) and describes using dropout with data augmentation for its ImageNet model (H Models for ImageNet, p. 18). It does not establish that YOLO uses either technique or describe YOLO’s augmentation as extensive. (LLM, conf=0.88)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“However, only 30Hz DPM [31] actually runs in real-time.”
What the sources say
- Supportedcites [31] [31] M. A. Sadeghi and D. Forsyth. 30hz object detection with dpm v5. In Computer Vision-ECCV 2014 , pages 65-79. Springer, 2014. 5, 6 - source
The Introduction (p.1) states that 'except for certain objects including face, pedestrians and certain rigid objects, detectors do not currently run at video rate' and that the presented architecture detects the 20 PASCAL VOC categories simultaneously at 30Hz, i.e. at video/real-time rate. The Abstract (p.1) and Table 2 (Method) bound the comparison to other DPM speed-up efforts: reference DPM at 0.07Hz, the fastest GPU 'Vanilla DPM' at ~1Hz, FTVQ at 0.53s, and 'over an order of magnitude faster than the previous fastest DPM implementation', so none of the listed variants reaches real-time. Thus our copy supports both parts of the attribution: that this 30Hz DPM runs in real-time and that it is the only such DPM among the speed-up efforts compared. (LLM, conf=0.82)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals.”
What the sources say
- Could not verifycites [20] [20] K. Lenc and A. Vedaldi. R-cnn minus r. arXiv preprint arXiv:1506.06981 , 2015. 5, 6 - sourceExperimental — accuracy not yet measured
The ConvNet-based Object Detection section identifies R-CNN minus R and says it uses predefined regions in an SPPnet system, but gives no speed comparison with R-CNN. A search for R-CNN minus R and speed found no passage in our copy establishing that it is much faster. (LLM, conf=0.73)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“While it is much faster than R-CNN, it still falls short of real-time and takes a significant accuracy hit from not having good proposals.”
What the sources say
- Partly supportedcites [20] [20] K. Lenc and A. Vedaldi. R-cnn minus r. arXiv preprint arXiv:1506.06981 , 2015. 5, 6 - sourceExperimental — accuracy not yet measured
The “ConvNet-based Object Detection” section (page unavailable) says R-CNN minus R uses predefined regions instead of region proposals. It does not say that this causes a significant accuracy loss; the paper’s discussion of low accuracy concerns a different system, naïve Faster R-CNN. (LLM, conf=0.83)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
84 more claims
Create a free account to see all 121 claims and to verify the 2 references still open.
- Probably not a substantial claim. 17
- It may state two separate claims as one. 15
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“which was not thoroughly investigated in prevIoUs work such as YOLOv1 [20] and YOLOv2 [51] employing MSE/L2 loss (Mean Squared Error).”
From Object Detection in 20 Years: A Survey (2023) · cites this paper as [20]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“the YOLO series models v2-v5 [144-147] are some of the most advanced object detection methods based on DNN, which have good performance in terms of speed and precision.”
From A survey of model compression strategies for object detection (2023) · cites this paper as [144]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“As a typical and more widely used new form of applications [9], various deep learning-based intelligent services and applications have changed many aspects of people's lives due to the great advantages of Deep Learning (DL) in the fields of Computer Vision (CV) and Natural Language Processing (NLP) [10].”
From Convergence of Edge Computing and Deep Learning: A Comprehensive Survey (2020) · cites this paper as [9]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“A convolutional neural network predicts the bounding boxes as well as class probabilities for all the objects depicted in an image.”
From Object detection using YOLO: challenges, architectural successors, datasets and applications (2023) · cites this paper as [56]
The cited source says what the claim attributes to it.
In the citing paper
“Authors of YOLO [56] have reframed the problem of object detection as a regression problem instead of classification problem.”
From Object detection using YOLO: challenges, architectural successors, datasets and applications (2023) · cites this paper as [56]
6 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
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
Self-citationFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
Self-citationGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
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 1×
“We pretrain our convolutional layers on the ImageNet 1000-class competition dataset [29].”
MethodHistograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · cited 2×
“Detection pipelines generally start by extracting a set of robust features from input images (Haar [25], SIFT [23], HOG [4], convolutional features [6]).”
MethodRich 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 · cited 2×
“Then, classifiers [35, 21, 13, 10] or localizers [1, 31] are used to identify objects in the feature space.”
MethodSelf-citationGirshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105
Self-citationDelving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.123 · Cited by 19,741 · cited 1×
“Convolutional neural networks (CNNs) achieve impressive performance on classification tasks at real-time speeds [13].”
Object recognition from local scale-invariant features
Lowe, D. G. (1999) · Proceedings of the Seventh IEEE International Conference on Computer Vision · doi:10.1109/iccv.1999.790410 · Cited by 16,472 · cited 1×
“Detection pipelines generally start by extracting a set of robust features from input images (Haar [25], SIFT [23], HOG [4], convolutional features [6]).”
MethodRobust Real-Time Face Detection
Viola, P., & Jones, M. J. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000013087.49260.fb · Cited by 14,097 · cited 1×
“Detectors for single classes like faces or people can be highly optimized since they have to deal with much less variation [36].”
Background
Cited by
Items 34,401–34,500 of 51,399
Mahdavian, M., Loni, M., Samuelsson, T., & Chen, M. (2025) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros60139.2025.11246817 · Cited by 2
Jayathunga, D. P., Ramashini, M., Zaini, J., Müller, R., & Silva, L. C. D. (2025) · Applied AI Letters · doi:10.1002/ail2.70014 · Cited by 1
Xiao, 冯. F., Weiwei, 龚. G., & Xiang'e, 孙. S. (2025) · Laser & Optoelectronics Progress · doi:10.3788/lop250854 · Cited by 1
Khairy, M., Talaat, A. S., & Al‐Makhlasawy, R. M. (2025) · Journal of Electrical Systems and Information Technology · doi:10.1186/s43067-025-00295-w · Cited by 1
“Scenes, which are composed of related shots, usually reflect a shift in time or location [9].”
AUP-DETR: A Foundational UAV Object Detection Framework for Enabling the Low-Altitude Economy
Xu, J., Liu, X., Li, X., & Hu, Y. (2025) · Drones · doi:10.3390/drones9120822 · Cited by 1
Substation Instrument Defect Detection Based on Multi-Domain Collaborative Attention Fusion
Liu, K., Li, Y., Wang, S., Yang, Z., Li, Z., & Zhao, Z. (2025) · Electronics · doi:10.3390/electronics14234690 · Cited by 1
YOLOv10-C2f-FENG: Efficient Anomaly Detection for Smart Care Monitoring
Xu, F., Lee, J.-C., & Pan, Y.-J. (2025) · doi:10.1109/iccbe65177.2025.11256033 · Cited by 1
“Two main categories exist for posture analysis: (1) keypoint ‑ based pose estimation (e.g., OpenPose [3]) that models human skeletons via part affinity fields, and (2) object ‑ detection ‑ based methods such as YOLO [4] that frame human bodies or posture groups as bounding boxes.”
Maghade, P. S., Naik, S., Koshti, S., Navale, R., Bhosale, A. G., & Harikrishnan, R. (2025) · International Conference on Recent Innovation in Science Engineering and Technology (ICRISET) · doi:10.1109/icriset64803.2025.11252224 · Cited by 1
Sable, N. P., Shukla, V. K., Khedkar, V. N., et al. (2026) · Frontiers in Computer Science · doi:10.3389/fcomp.2026.1855851 · Cited by 1
Data rescue of historical tables through semi-supervised table structure recognition
Singh, L. G., & Middleton, S. E. (2025) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-025-00562-6 · Cited by 4
“Popular object detection models like Faster R-CNN [23], RetinaNet [24], and YOLO [25], alongside segmentation models such as TableNet [26], Cas-cadeTabNet [18], have been adeptly adapted for TSR task.”
Liu, D., Xiao, Y., Liu, Y., He, J., Liu, K., Yu, D., Liu, L., Chen, Y., Gong, X., & Liu, J. (2025) · International Journal of Applied Ceramic Technology · doi:10.1111/ijac.70122 · Cited by 1
A high precision and lightweight method for steel surface defect detection based on improved YOLOv5
Zhou, M., Wang, H., & Wang, Y. (2025) · Scientific Reports · doi:10.1038/s41598-025-28022-5 · Cited by 1
ISO-DeTr: A novel detection transformer for industrial small object detection
Saeed, F., & Paul, A. (2025) · Machine Learning with Applications · doi:10.1016/j.mlwa.2025.100809 · Cited by 3
Low Altitude UAV Detection in Complex Contexts
Tang, Z., Lu, W., Tang, W., et al. (2025) · IEEE Access · doi:10.1109/access.2025.3639379 · Cited by 2
Clyburn, M., Nasir, N., Billah, M. F. R. M., Sobral, V. A. L., Gao, J., Nikseresht, F., & Campbell, B. (2025) · ACM/IEEE Symposium on Edge Computing (SEC) · doi:10.1145/3769102.3770617 · Cited by 1
“…a loop, matrix multiplication multiplies two matrices, Fast Fourier Transform computes an FFT, human activity recognition uses a neural network for gesture recognition, object detection uses YOLO [27] to classify objects in an image, and room classification uses SqueezeNet [17] to classify a room.”
LPC-SYS: AI-Powered Laryngo-Pharyngeal Cancer Diagnosing System
Li, Y., Chen, H., Zhou, P., et al. (2025) · doi:10.1109/embc58623.2025.11251675 · Cited by 1
“In contrast, single-stage detection methods, including YOLO [9], SSD [10] and RetinaNet [11] offer a simpler model and faster analysis speed, making them particularly suitable for tasks with high real-time requirements.”
Evaluating Efficient Patch-Based Backdoor Attacks in Satellite Image Classification Systems
Rahmanian, G., & Madani, P. (2025) · International Conference on Privacy, Security and Trust (PST) · doi:10.1109/pst65910.2025.11268845 · Cited by 1
“Architectures such as ResNet [2] and YOLO [3] have become popular foundations for different computer vision purposes due to their proven accuracy and versatility across a wide range of visual classification tasks.”
Fine-grained feature enhancement and occlusion-aware dense pedestrian detection in complex scenes
Li, M., & Huang, S. (2025) · Multimedia Systems · doi:10.1007/s00530-025-02083-y · Cited by 3
“The YOLO family has evolved significantly, progressing from YOLOv1 [4] in 2016 to the latest YOLOv12 [12].”
Deep Learning Framework for Multiclass Segmentation of Photovoltaic Systems
Krikau, S., & Keller, S. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3640268 · Cited by 2
Li, L., Jiang, Q., & Bai, D. (2025) · Engineering Research Express · doi:10.1088/2631-8695/ae2834 · Cited by 1
ADF-YOLO: lightweight adaptive network for high-speed pharmaceutical blister pack detection
Yu, Y., Zhou, P., Fu, H., & Tang, C. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01809-5 · Cited by 2
“YOLOv1 [22] initially proposed an end-to-end detection framework and utilized a single-stage network architecture to process images, whereas YOLOv3 introduced a multi-scale prediction mechanism; this series of algorithms gradually constructed the theoretical foundation of real-time target detection.”
Real-Time Sugarcane Bud Detection Using Efficient Lightweight Convolutional Blocks
Kumar, P., Arora, S., & Arora, S. (2025) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-025-10906-3 · Cited by 1
Xu, M., Dai, L., Wu, H. R., & Chen, Z. (2025) · The Visual Computer · doi:10.1007/s00371-025-04289-7 · Cited by 1
“Kim et al.[10] used video data and combined the YOLOv4 object detection model with the DeepSORT tracking algorithm to count pigs in corridor videos, achieving an accuracy of 99.4% on counting with adult pigs.”
InfluentialImproved YOLOv5s-Based Crack Detection Method for Sealant-Spraying Devices
Kong, W., Ding, H., Cheng, Q., Li, N., Sun, X., & Dong, X. (2025) · Symmetry · doi:10.3390/sym17122089 · Cited by 1
Usta, D. D., Rubio, R. H., Martínez, C. L., et al. (2025) · Future Transportation · doi:10.3390/futuretransp5040191 · Cited by 1
Yolov8n-RCP: An Improved Algorithm for Small-Target Detection in Complex Crop Environments
Xing, J., Yan, H., Li, Z., Zhu, J., Zhang, L., & Zhang, L. (2025) · Electronics · doi:10.3390/electronics14244795 · Cited by 2
Li, J., Bitter, K., Nguyen, A. D., Shemesh, H., Zaslansky, P., & Zachow, S. (2025) · Dentistry Journal · doi:10.3390/dj13120579 · Cited by 1
“Hasan, H.A. et al. [75] used the YOLO network [109] to classify the post-operative X-rays into different outcome groups, such as suboptimal treatment, incomplete or complete obturation.”
Extension of Intersection Over Union to Improve Small Object Detection in Few-Shot Regime
Jeune, P. L., & Mokraoui, A. (2025) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco63237.2025.11226402 · Cited by 1
“Most detection pipelines, including few-shot object detection (FSOD), rely on Intersection over Union (IoU) as a regression loss [8], [9], for example selection [1], [2], [10], or as an evaluation metric.”
S-RRT: A Semantic-Driven Extension of the Rapidly-Exploring Random Tree Algorithm
Nguyen, V. H., Pham, V. M., Nguyen, V. T., & Truong, X. T. (2025) · International Conference on Advanced Technologies for Communications · doi:10.1109/atc67618.2025.11268639 · Cited by 1
“By leveraging AI, methods such as semantic segmentation [20] and object detection [21] have achieved significant improvements in both performance and accuracy.”
Xiao, B., Wen, J., He, X., Wu, Y., Zhang, X., & Hu, S. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01816-6 · Cited by 1
“These methods are primarily categorized into two-stage detectors [3] (e.g., Faster R-CNN [4]) and one-stage detectors (e.g., YOLO [5], RetinaNet [6]).”
Kim, J. E., & Choi, D. (2025) · Journal of the Korea Institute of Military Science and Technology · doi:10.9766/kimst.2025.28.6.564 · Cited by 1
“[4] 에 의해 처음 제안된 이후 , Ultralytics 의 YOLOv5 [15] , YOLOv8 [6] , YOLOv9 [16] 등으로 발전하였다 .”
MSTA-YOLOv11: an improved detection model for maize pests and diseases under complex scenes
Yuan, L., & Wang, F. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01821-9 · Cited by 1
“Detection frameworks such as Faster R-CNN [9], SSD [10], and YOLO [11] have been applied to crop pest and disease detection.”
Confidence-Guided Code Recognition for Shipping Containers Using Deep Learning
Khuboni, R., Hlabisa, S., & Tapamo, J. (2025) · Big Data and Cognitive Computing · doi:10.3390/bdcc9120316 · Cited by 1
Zheng, B., Angkawisittpan, N., Huang, L., et al. (2025) · Engineering Technology & Applied Science Research · doi:10.48084/etasr.14220 · Cited by 1
Review of Aneurysms Detection Methods Focusing on Selected YOLO-Based Models
Kamencay, P., Hlavata, R., Paralič, M., et al. (2025) · Journal of Clinical Medicine · doi:10.3390/jcm14248716 · Cited by 1
“A major advantage of the YOLO architecture is its ability to process the entire image in a single pass during prediction, enabling it to leverage global image context [17].”
Hmimou, Y., Hidila, Z., & Khiat, A. (2025) · IEEE International Conference on Electronics, Circuits and Systems (ICECS) · doi:10.1109/icecs66544.2025.11270781 · Cited by 1
“The advent of deep learning, and especially the introduction of the You Only Look Once (YOLO) framework [6], has revolutionized real-time object detection and significantly influenced ALPR research.”
Wu, X., Yang, Z., Hou, C., Huo, D., Ge, Y., Long, Q., Yin, P., & Luo, X. (2025) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-025-01767-1 · Cited by 1
“Sin-gle-stage Sin-gle-stage instance segmentation models simultaneously perform object detection and pixel-level segmentation in a single forward pass, emphasizing speed and efficiency, such as SSD [15] and YOLO [16].”
Gao, G., & Chen, C. (2025) · Textile & Leather Review · doi:10.31881/tlr.2025.925 · Cited by 2
Toward Efficient Semi-Supervised Object Detection With Detection Transformer
Zhang, J., Li, J., Lin, X., Zhang, W., Tan, X., Gao, H., Wang, J., & Li, G. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3642123 · Cited by 2
“In contrast, one-stage detectors [60], [61], [62], [73], [97] streamline the process by directly predicting bounding boxes and class probabilities from image features, eliminating the proposal generation step.”
Who Is Smoking? A Frame Analysis-Based Cigarette Detection Framework for Offline Videos
Koyun, O. C., Keser, R. K., & Töreyın, B. U. (2025) · EAI/Springer Innovations in Communication and Computing · doi:10.1007/978-3-031-92143-8_9 · Cited by 1
Robotic Masonry Arch Construction Considering Object Recognition and Human-Robot Interaction
Koliyaee, V., Lin, X., Liu, R., et al. (2025) · ASCE International Conference on Computing in Civil Engineering (i3CE) · doi:10.1061/9780784486115.073 · Cited by 1
Nguyen, L., Nguyen, B. H., Walker, S. P., et al. (2025) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz67716.2025.11281848 · Cited by 1
“Real-time object detection models such as YOLO have also improved large-scale, rapid screening, with YOLOv10 [12] showing promising results for fish spinal deformities [7], [13].”
Beyond Deep Learning: Agentic AI Framework for Object Detection
Shah, S. A. A. (2025) · International Conference on Image and Vision Computing New Zealand (IVCNZ) · doi:10.1109/ivcnz67716.2025.11281644 · Cited by 1
“You Only Look Once (YOLO) [17] reframed object detection as a regression problem, allowing real-time performance but initially sacrificing localization accuracy.”
Seafood Object Detection Method Based on Improved YOLOv5s
Zhu, N., Liu, Z., Wang, Z., et al. (2025) · Sensors · doi:10.3390/s25247546 · Cited by 1
“In contrast, one-stage detectors like the YOLO series [14–19], SSD [20], and RetinaNet [21] perform classification and localization in a unified pipeline, offering improved inference speed suitable for real-time applications.”
Ni, Z., & Kim, Y. (2025) · Applied Sciences · doi:10.3390/app152413115 · Cited by 5
A Scalable Video Analysis Framework for Character Identification and Statistical Profiling
Liu, C.-C., & Lin, Q.-F. (2025) · IEEE Global Conference on Consumer Electronics (GCCE) · doi:10.1109/gcce65946.2025.11274842 · Cited by 1
“The system initializes by verifying dependencies (MediaPipe [2], FaceNet [1], YOLOv3 [3]) and performing video integrity checks.”
IDSE-YOLO: An Infrared Recognition Model for Substation Equipment
Ding, H., & Zheng, J. (2025) · IEEE Access · doi:10.1109/access.2025.3643499 · Cited by 3
Siamese CNN-based Few-shot Learning for PCB Defect Detection*
Chiu, H.-C., Hsiao, C.-H., & Wang, Y.-T. (2025) · IEEE Global Conference on Consumer Electronics (GCCE) · doi:10.1109/gcce65946.2025.11274903 · Cited by 1
“Traditional deep learning models, such as those in the YOLO series, deliver impressive results— particularly when trained on large, well-balanced datasets [2].”
Li, Z., Wang, Y., Meng, L., Chu, W., & Yang, G. (2025) · Journal of Imaging · doi:10.3390/jimaging11120447 · Cited by 4
“Single stage detectors such as YOLO series [3] and SSD [4] achieve fast detection through end-to-end architecture and are suitable for tasks with strong real-time requirements.”
139Chapter 8 Augmented realities: AI revolutionizing Industry 4.0 automation with IoT integration
Gupta, A. (2025) · De Gruyter eBooks · doi:10.1515/9783111388588-008 · Cited by 1
Hou, S., Wang, B., Wang, N., Hou, J., & Pang, Y. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01818-4 · Cited by 3
“Deep learning has advanced object detection in satellite remote sensing, with two-stage (R-CNN [6], Faster R-CNN [7]) and single-stage (YOLO [8], SSD [9]) paradigms.”
YOLO-VML: An Improved Object Detection Model for Blastomeres and Pronuclei Localization in IoMT
Shen, A., Li, C., Yang, J., Huang, G., & Zhang, X. (2025) · IEEE Open Journal of Engineering in Medicine and Biology · doi:10.1109/ojemb.2025.3644699 · Cited by 1
“Theemergenceofobjectdetectionalgorithms basedontheyouonlylookonce(YOLO)series[9],[10],[11], [12],[13],[14],[15],[16],[17]hassignificantlyadvancedthe developmentofthisfield.”
AI-Driven Optical Intelligence for Modern Warfare Command and Control
Chougule, R., Patil, V., Rane, A., Mangla, M., Hole, P., & Sankhe, M. (2025) · International Conference on Computing Communication Control and automation · doi:10.1109/iccubea65967.2025.11283928 · Cited by 1
“Most related studies focus on object detection using convolutional neural networks (CNNs) [5] and feature extraction techniques such as YOLO [6] and Faster R-CNN [7].”
Hybrid Deep Learning Framework for Long-Term Forecasting of PV Generation
Babic, L., & Aunedi, M. (2025) · International Universities Power Engineering Conference (UPEC) · doi:10.1109/upec65436.2025.11279840 · Cited by 1
“Convolutional neural networks (CNNs), first introduced for digit classification [6], and subsequently proven effective across different image classification applications [27, 28], excel at detecting local patterns that occur at adjacent input elements (i.e., pixels).”
Hands-on Evaluation of Visual Transformers for Object Recognition and Detection
Vlachogiannis, D. N., & Koutsomitropoulos, D. A. (2025) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai66417.2025.00125 · Cited by 1
“Ar-chitectures such as AlexNet [1], VGGNet [2], ResNet [3] for object recognition and Faster R-CNN [4], Yolo [5] for object detection have established CNNs as the dominant architecture in computer vision.”
Suslenko, O. (2025) · Cybernetics and Computer Technologies · doi:10.34229/2707-451x.25.4.9 · Cited by 1
RCAN-DETR: a PCB defect detection model based on re-calibrated multi-scale fusion
Ling, W., Li, X., Chen, K., Chen, F., & Song, X. (2025) · Engineering Research Express · doi:10.1088/2631-8695/ae2ceb · Cited by 2
CsDETC: detection and counting of small target Cryptococcus spp.
Chen, Y., Luo, Y., Yang, Z., Hu, L., Zhong, Q., & Sheng, T. (2025) · Frontiers in Cellular and Infection Microbiology · doi:10.3389/fcimb.2025.1701899 · Cited by 1
SARPAR: Systolic ARray Pallet-Integrated AcceleratoR for YOLO models on FPGA
Eydivandi, S., & Beitollahi, H. (2025) · Integration · doi:10.1016/j.vlsi.2025.102634 · Cited by 2
Real-time hybrid AI for strip steel surface monitoring in industry 4.0
Sabri, K., & Gaceb, M. (2025) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-025-17113-2 · Cited by 1
“The YOLOv8n architecture was adopted for its balance between accuracy and efficiency in real-time inspection tasks [39 –41].”
Ram, B. G., Joy, J., Marcotte, N., Urlacher, S., McDonald, D., Amundson, J. K., Silewski, G., Jhala, A. J., & Sun, X. (2025) · Journal of Agriculture and Food Research · doi:10.1016/j.jafr.2025.102577 · Cited by 5
Using YOLOv8 for Milkfish (Chanos chanos) Fry Counting in Philippine Hatcheries
Espeña, G. D., Casarungan, A. I. F. D., Villaverde, O. S. M., et al. (2025) · IEEE International Conference on Internet of Things and Intelligence Systems (IoTaIS) · doi:10.1109/iotais67227.2025.11282093 · Cited by 1
Mai, R. (2025) · International Conference on Artificial Intelligence, Digital Media Technology and Social Computing (ICAIDS) · doi:10.1109/icaids67687.2025.00014 · Cited by 1
“These algorithms are commonly employed in real-time detection scenarios, with representatives including SSD [5] , YOLO [6-9] , and RetinaNet [10] .”
Urban Flow - A Integrated Smart Traffic Management System
Patil, S., Baji, O., Bagul, M., et al. (2025) · Asian Conference on Innovation in Technology (ASIANCON) · doi:10.1109/asiancon66527.2025.11281127 · Cited by 1
“YOLO, a single deep neural network for real-time object detection, redefines detection as one regression task, predicting bounding boxes and class probabilities directly from whole images [5].”
Faster R-CNN Marine Benthos Detection with Multi-scale Channel Attention Mechanism and RoIAlign
Qiu, Y., Zhou, Z., & Yang, J. (2025) · Arabian Journal for Science and Engineering · doi:10.1007/s13369-025-10962-9 · Cited by 1
“Examples include SSD series [16], DSSD series [17], and the YOLO series [18, 19].”
A Symmetry-Aware Hierarchical Graph-Mamba Network for Spatio-Temporal Road Damage Detection
Tian, Z., Shao, X., Bai, Y., et al. (2025) · Symmetry · doi:10.3390/sym17122173 · Cited by 1
“This limitation spurred the development of one-stage detectors, with the YOLO series [2–4] and the Single Shot Detector (SSD) [9] emerging as the most prominent paradigms.”
Zhu, Y., Zhu, W., & Wang, L. (2025) · International Conference on Artificial Intelligence, Automation and High Performance Computing (AIAHPC) · doi:10.1109/aiahpc66801.2025.11290184 · Cited by 1
“Since YOLOv1, the YOLO series has advanced steadily [12] .”
Small-target traffic sign detection based on improved YOLOv8
Ma, Y., Chen, L., Gao, B., Li, Y., Liu, S., & Liu, Z. (2025) · Engineering Research Express · doi:10.1088/2631-8695/ae2f1d · Cited by 1
Amp: single-shot ultra-wide fisheye-to-cubemap PnP pose estimation
Raettig, R. M., Nyquist, R. R., Nykl, S. L., Taylor, C. N., & Kabban, C. M. S. (2025) · Machine Vision and Applications · doi:10.1007/s00138-025-01768-8 · Cited by 1
“At the Air Force Institute of Technology (AFIT), ongo - ing AAR computer vision pipeline research [5, 46 ] employs You Only Look Once (YOLO) [ 36 ] to detect and locate cus - tom classes on a known 3D rigid object within an image.”
Zhuo, S., Zhou, X., Cai, J., Bai, H., Duan, X., & Ma, Y. (2025) · PLoS ONE · doi:10.1371/journal.pone.0336971 · Cited by 1
“You Only Look Once (YOLO) [24] is a real-time object detection algorithm that predicts object categories and bounding boxes in a single forward pass, achieving end-to-end detection.”
Praneeth, Raj, A., Kunal, K., et al. (2025) · doi:10.1109/csitss67709.2025.11295736 · Cited by 1
“Redmon et al. [1] introduced the YOLO (You Only Look Once) framework, reframing object detection as a single regression problem, enabling real-time detection with competitive accuracy.”
MSH-YOLO: Multi-scale Hierarchical Adaptive Network for Small Object Detection in UAV Aerial Images
Guo, S., Lu, X., & Wang, Y. (2025) · doi:10.1109/leficlu65987.2025.11297274 · Cited by 1
“In 2015, R. Joseph et al.[9] proposed the single-stage detection algorithm YOLO, which applies a single neural network to entire images, dividing images into grids and simultaneously predicting bounding boxes and classes for each grid.”
Zhao, Y., & Qian, H. (2025) · Scientific Reports · doi:10.1038/s41598-025-33107-2 · Cited by 3
Enhancing low-light RGB-D pedestrian detection through dual-stage modality-guided fusion
Guo, J., & Sun, G. (2025) · The Visual Computer · doi:10.1007/s00371-025-04237-5 · Cited by 1
Bai, T., Wang, X., Wang, Y., Li, W., & Xiang, H. (2025) · The Journal of Supercomputing · doi:10.1007/s11227-025-08155-0 · Cited by 1
“The You Only Look Once (YOLO) [27] model series is a deep learning-based object detection framework widely used for railway track defect detection.”
Infrastructure-based sensor fusion for acquiring gross vehicle weight rating classifications
Feng, G., Li, Y., Tok, A. Y. C., & Ritchie, S. G. (2025) · Transportation Research Interdisciplinary Perspectives · doi:10.1016/j.trip.2025.101535 · Cited by 1
Practical Mobile Based Services for Identification of Chicken Diseases from Fecal Images
Silapachote, P., Srisuphab, A., Damkham, W., et al. (2024) · IEEE Region 10 Conference (TENCON) · doi:10.1109/tencon61640.2024.10902790 · Cited by 1
Artificial intelligence for remote sensing and climate monitoring
Tang, H., Bhatti, U. A., Li, D., Lisi, D., Liu, J., & Bhatti, M. A. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-24712-5.00004-x · Cited by 2
WT-YOLO: A High-Accuracy Model for Wind Turbine Target Detection
Gu, Z., Peng, Y., & Sun, C. (2025) · Communications in computer and information science · doi:10.1007/978-981-96-9964-3_15 · Cited by 1
Jiang, X., Zeng, W., Lu, J., et al. (2025) · Marine Energy Research · doi:10.70322/mer.2025.10013 · Cited by 1
MAD-Flow: An Efficient Deployment Flow for Memory-Bank-Based Anomaly Detection on FPGA-SoCs
Wu, W., Chen, C., Wu, Y., & Wang, H. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3605880 · Cited by 2
“The first, and most dominant, area is the acceleration of supervised object detection models, such as the YOLO [14] family, for defect detection tasks [15], [16], [17].”
Capuano, G. M., Capuozzo, S., Strollo, A. G. M., & Petra, N. (2025) · Acta Astronautica · doi:10.1016/j.actaastro.2025.09.018 · Cited by 6
Wilson, K. C., Lurbur, M., & Yochum, N. (2025) · ICES Journal of Marine Science · doi:10.1093/icesjms/fsaf168 · Cited by 3
Infantile Spasm Detection Based on Spatio-Temporal Attention Graphconv U-Net
Meng, L., Wang, Y., & Fan, Y. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3614418 · Cited by 2
Yuan, J., Ma, W., Han, M., Nian, W., & Gao, L. (2025) · The Journal of Engineering · doi:10.1049/tje2.70123 · Cited by 1
Efficient Multiaspect Compensation Network for Small Object Detection in SAR Images
Li, C., Zhang, P., Hei, Y., et al. (2025) · IEEE Transactions on Aerospace and Electronic Systems · doi:10.1109/taes.2025.3615169 · Cited by 1
“Seeking to balance speed and precision, researchers have introduced one-stage detectors, such as you only look once (YOLO) series [22], [23], [24].”
Şımşek, E., & Özyer, B. (2025) · Image and Vision Computing · doi:10.1016/j.imavis.2025.105742 · Cited by 2
Pediatric Dental Disease Detection Using X-Ray Image Enhancements and Deep Learning Algorithms
Jaidev, K., Vignesh, M., Raveena, R., et al. (2025) · Springer series in reliability engineering · doi:10.1007/978-3-031-98728-1_5 · Cited by 1
Shi, W., & He, Y. (2025) · Springer eBooks · doi:10.1007/978-3-031-99485-2_3 · Cited by 1
Smart Drones in Search and Rescue: Leveraging AI for Autonomous Decision-Making
Alsaedi, N., Alomari, E., Alseiari, H., et al. (2025) · International Conference on Activity and Behavior Computing (ABC) · doi:10.1109/abc64332.2025.11118531 · Cited by 1
“In this paper, we employ You Only Look Once (YOLO) model introduced by J. Redmon et al. [21].”
Advanced PPE Compliance with AI Driven Insights
Patil, S., Sandve, A., Wanjari, D., Waghmode, N., & Bagade, S. (2025) · International Conference on Intelligent Technologies (CONIT) · doi:10.1109/conit65521.2025.11167141 · Cited by 2
“[1] YOLO is fast because it processes the image through the network to produce the final detection result, enabling real-time video detection as well[3].”
Hermann, J., Topac, Y., Hoffmann, J., Wedemeyer, L., Jansen, N., & Doğangün, A. (2024) · Australian Conference on Human-Computer Interaction (OzCHI) · doi:10.1145/3726986.3727026 · Cited by 2
Design and Development of a Real-Time Camera-Based Smart Cooking Assistant
Sheikh, H., George, K., Nobari, T. Z., & Panangadan, A. V. (2025) · IEEE International Conference on Information Reuse and Integration (IRI) · doi:10.1109/iri66576.2025.00073 · Cited by 1
“Unlike traditional object detection methods, YOLO uses a single unified network to simultaneously detect and classify objects, greatly simplifying the detection pipeline compared to earlier approaches [15].”
A Lightweight Image Segmentation Algorithm Based on YOLOv8
Zehua, W., Kong, S., & Junzhi, Y. (2025) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac66630.2025.11150188 · Cited by 1
“[11].”
Automated MLOps-Driven YOLO Framework for Drone-Based Plant Disease Detection
Elfeky, S. M. M., Yehia, M. N. A., & Hamdi, A. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-6441-2_35 · Cited by 1
MFJLN: Multi-Frequency Feature Joint Learning Network for Rain Removal
Yang, Y., Yang, J., Huang, S., & Wan, W. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3618579 · Cited by 1
YOLO-BT: A novel brain tumor detector with gradient-guided feature diffusion
Zhuang, D., Yao, Q., Feng, Y., Xv, H., Zhang, C., & Wang, X. (2025) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2025.108696 · Cited by 2
Fulin, 唐. T., Weishu, 李. L., & Guanjun, 乔. Q. (2025) · Laser & Optoelectronics Progress · doi:10.3788/lop250655 · Cited by 1
Plug-and-Play Shape Matching Module for Zero-Shot Mesh-Free Grasp Refinement on Unknown Objects
Hong, J. Y., Son, Y. G., Um, S. H., & Choi, H. R. (2025) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2025.3623004 · Cited by 2
“While classical semantic [28] and instance segmentation models [29], [30], [31] are powerful, their performance is tied to the classes seen during training.”
Dantas, M., Silva, M. L., Filho, A. T. D. O., Bezerra, D. D. F., Freitas, E., do Carmo, P. R. X., Sadok, D. F. H., Kelner, J., & Souza, R. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06916-1 · Cited by 1
“The shift towards using advanced deep learning models like pruned YOLO [113] or multi-exit architectures marks a significant evolution from traditional machine learning ● Edge Computing Integration: An exploration of the integration between DTs and edge computing paradigms could significantly…”
Metadata
- Authors
- Redmon, J., Divvala, S., Girshick, R., & Farhadi, A.
- Affiliations
- University of Washington (US)
- Meta (United States) (US)
- Venue
- 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2016-06
- Pages
- 779-788
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W1483870316
- Cited by
- 53,803
- References
- 42
- Citations by year
- 2026: 5347 · 2025: 8912 · 2024: 8953 · 2023: 8264 · 2022: 6785 · 2021: 5741 · 2020: 4223 · 2019: 3286 · 2018: 1560 · 2017: 635 · 2016: 90
- FWCI
- 644.23
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Zenodo (CERN European Organization for Nuclear Research) (cc-by)
- Funders
- National Science Foundation (1338054)
- Office of Naval Research (N00014-13-1-0720, N00014)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- object detection, YOLO, real-time detection, bounding box regression, end-to-end learning, false positive reduction, domain generalization
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
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.