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 50,201–50,300 of 51,398
Robust Low-Light Pedestrian Detection via Optimized Zero-DCE and YOLOv8
Zhang, D., Yue, E., Liu, B., & Wu, D. (2026) · IEEE Access · doi:10.1109/access.2026.3719936
Han, J., Wang, Z., Liu, W., Song, B., Wang, C., & Liu, X. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116784
Akylbekov, O., Omirzak, A. A., & Moldagulova, A. (2026) · International Conference on Control and Automation · doi:10.1109/icca69928.2026.11617995
An Improved YOLOv8 Structure for Multi-Scale Object Detection in UAV Aerial Imagery
Zeng, Y., Zhong, Q., Fang, Z., et al. (2026) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac71005.2026.11615360
“As a representative one-stage detector, the YOLO series is widely used due to its favorable speed-accuracy trade-off and has continuously evolved in aspects like multi-scale feature fusion, loss design, and training strategies [6] .”
Autonomous Aerial Delivery of Flapping-Wing Flying Robots
Liu, C., Zhang, X., Meng, T., Wu, X., & He, X. (2026) · Youth Academic Annual Conference of Chinese Association of Automation (YAC) · doi:10.1109/yac71005.2026.11615326
“Although the YOLO series achieves strong detection performance [22], [23], baseline models are often unstable under FWFR vibration and motion blur.”
AI-based Automated Food Recognition and Pay-As-You-Eat Billing System
Musqaty, M. B. A. A., Musqaty, A. R. B. A. A., Moizuddin, S. M., et al. (2026) · 2026 6th International Conference on Inventive Computation and Information Technologies (ICICIT) · doi:10.1109/icicit69063.2026.11634131
“Although invalid scans and deliberate obstructions can impact billing accuracy or system reliability, very little has been written about their detection using scanning on food trays [12].”
InfluentialChowdhury, N. H., Rahim, M., Hasan, S. F., Hasan, M., & Bhattacharjee, P. (2026) · Neural computing & applications (Print) · doi:10.1007/s00521-026-12329-z
“For the ability to be able to reproduce the same result with consistency, these models were trained using the Ultralytics YOLO framework.”
SortWaste: A Densely Annotated Dataset for Object Detection in Industrial Waste Sorting
Inácio, S., Proença, H., & Neves, J. C. (2026) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw68408.2026.00065
“YOLO family of detectors [23] formulates object detection as a single-stage regression problem, enabling real-time performance.”
ASAS: Adaptive Sample Assignment Strategy for Single-Head Object Detection
Mboutayeb, S., Majda, A., & Zenkouar, K. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-18860-1_30
Park, S., Kim, G., Song, Y., Tian, C., Baek, C., Cho, Y., Kim, J. W., Jin, E., & Cho, S. (2026) · Analytical Chemistry · doi:10.1021/acs.analchem.6c02572
Real-Time Hazard Detection using an AI-Enabled Advanced Driver Assistance System
Joshika, P., Dharshana, C., Sridharan, S., et al. (2026) · doi:10.1109/icicit69063.2026.11634276
“These improvements put AI in a position to play an important role in enhancing a driver’s ability to be aware of his/her environment and make timely decisions, thus decreasing the chances of being involved in a traffic accident [1].”
Advancing Robust Infrared Object Detection: Practical Insights on Model Development and Evaluation
Stadler, D. (2026) · 2026 IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw68408.2026.00022
“The YOLO family [1, 7–9, 20–22, 24, 25, 27, 28] has emerged as the predominant framework for real-time detection.”
Wang, Y., Zou, Y., Wu, D., Zhu, X., & Xu, T. (2026) · Mechanisms and machine science · doi:10.1007/978-3-032-28967-4_55
Object Detection and Tracking: A Hybrid Approach Based on Deep Learning and Classical Algorithms
Kadam, P. P., Kale, L., Sawant, V. G., Khilari, A., & Balkrishnan, B. (2026) · International Conference on Data Science for Cyber-Physical Systems Resilience using Advanced Applications (ICDCA) · doi:10.1109/icdca69396.2026.11620279
Trust-Guided Multimodal LLM Integration with Reinforcement Learning for Autonomous Driving
Chennaka, S., & Nidamanuri, J. (2026) · IEEE/CVF Winter Conference on Applications of Computer Vision Workshops (WACVW) · doi:10.1109/wacvw68408.2026.00192
Liu, C., Shao, S., Cai, Z., et al. (2026) · AgriEngineering · doi:10.3390/agriengineering8080323
“For data annotation, the LabelImg tool was used to manually label individual wheat grains in each image using rectangular bounding boxes [27].”
EG-YOLO: a lightweight and high-efficiency network for real-time detection of steel surface defects
Chen, Z., & Ni, D. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01947-4
“Current deep learning detectors are mainly two-stage (e.g., Faster R-CNN) and one-stage (e.g., YOLO [8], SSD [9]) architectures.”
PCFD-Net: A Parallel Collaborative Fusion-Detection Network for SAR and Optical Imagery
An, Y., Wang, N., Wu, H., Wu, Y., & Liu, T. (2026) · Remote Sensing · doi:10.3390/rs18152595
SRC-Conv: Statistical Recalibration Convolution for Amorphous Fire and Smoke Detection
Ding, N., Sun, H., Luo, W., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31438-3_41
Yu, H., Xia, M., Tian, X., et al. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7237145
AS-YOLOR: An Improved YOLO Model for Small Object Detection in Aerial Images
Huang, J., Shen, J., Wang, M., & Zhou, Y. (2026) · Applied Sciences · doi:10.3390/app16157812
Current Research Status and Key Technological Advances of Refueling Robots
Zhou, S., Cui, W., Bai, W., Chen, S., Hua, W., Wu, Z., & Chen, Y. (2026) · Machines · doi:10.3390/machines14080892
LTE-DETR: Long-Tail Scenes Object Detection via Enhanced Feature Representation
Xu, F., Zhang, H., Zhang, Z., et al. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-8447-5_13
Che, Y., Zeng, Y., Ma, Z., Xia, Q., Zhang, R., Chen, L., Zhang, J., & He, H. (2026) · Sensors · doi:10.3390/s26154948
“Notably, object detection algorithms such as YOLO [8–11] and Faster R-CNN [12], alongside tracking algorithms like SORT [13], DeepSORT [14,15], and Byte-Track [16], have been successfully employed for dynamic species counting, including fish [17–19], sheep [5,20], chickens [21], and pigs [22–24].”
SFEDet: edge-guided spatial-frequency fusion network for steel surface defect detection
Chen, W., Wei, J., Guo, X., & Chen, B. (2026) · Scientific Reports · doi:10.1038/s41598-026-64588-4
A Gear Coupling Network for Infrared Dim and Small Target Detection in Complex Backgrounds
Zhu, B., Feng, X., Chen, Y., Xiaotong, Y., & Gaoxiang, H. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3720893
Rahmoune, H., Boutrid, N., & Benchoufi, I. (2026) · Artificial Intelligence in Gastroenterology · doi:10.35712/aig.v7.i2.116570
Barranco, F. R., Jo, K.-H., Cáceres-Hernández, D., et al. (2026) · Signals and communication technology · doi:10.1007/978-3-032-18566-2_17
Long‐Range Depth Estimation Using Learning‐Based Hybrid Distortion Model for CCTV Cameras
Pandat, A., Punna, R., Aravamuthan, G., Vinod, G., & Shukla, R. (2026) · Journal of Sensors · doi:10.1155/js/9911829
“For example, an efficient real-time detector such as YOLO [ 8] reliably detects drones with sub-pixel accuracy; but accurate localization is limited to short ranges (8 – 10m) [9].”
SenSNN: Energy-Efficient Tiny-Object Pseudo-Mask Segmentation via a Dual-Stream Membrane Shortcut
Wang, R., Li, H., & Guo, X. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3720766
Dumont, F., Raharjo, A. B., Dumas, C., Najem, C., Thibaudeau, E., Andyartha, P. K., & Vignaud, T. (2026) · European Journal of Surgical Oncology · doi:10.1016/j.ejso.2026.112045
Human motion-relied social interaction intention prediction for autonomous mobile robots
Bozorgi, H., & Ngo, T. D. (2026) · Transactions of the Canadian Society for Mechanical Engineering · doi:10.1139/tcsme-2025-0228
Hacıefendioğlu, K., Günaydın, M., Bostan, A., & Altunışık, A. C. (2026) · Sensors · doi:10.3390/s26154977
“Although two-stage object detection methods such as R-CNN and its derivatives [22,24] are known for their high accuracy, single-stage architectures have become increasingly popular in real-time applications due to their faster inference speed [23].”
Mahmood, F., Chryssinas, G., Inglezou, M., et al. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1881543
Chang, Y. L., Zhao, X., Li, Y., et al. (2026) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.70893
Estrella, D. R. Ñ., Meneses, J., & Sanmartín, H. (2026) · International Conference on Artificial Intelligence, Robotics and Control (AIRC) · doi:10.1109/airc69745.2026.11631485
“YOLO performs single-stage detection for real-time operation [1], with later designs improving the accuracy–speed trade-off [2].”
Real-Time Weapon Detection Using Enhanced YOLO12 Models and a Custom Dataset
Catargiu, C., & Ciocoiu, I. B. (2026) · IEEE Access · doi:10.1109/access.2026.3721157 · Cited by 1
Kong, Q.-Y., Yang, C., Tsai, C.-H., & Lee, M.-Y. (2026) · Journal of Hydrology · doi:10.1016/j.jhydrol.2026.136183
Dong, X., Shen, J., Li, C., & Zhang, T. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01955-4
“In contrast, single-stage methods (e.g., YOLO [6–8], SSD [9]) directly predict bounding boxes and categories, boasting a faster processing speed.”
Li, J., Wu, W., Che, W. G., et al. (2026) · Applied Sciences · doi:10.3390/app16157845
FUSEP: A Multi-Center Benchmark for Diverse Tasks in Fetal Ultrasound Screening Early Pregnancy
Pu, B., Yang, J., Wang, L., Tan, Y., He, G., Dong, X., Lin, Q., Guo, J., Yang, L., Liu, Z., et al. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3817453 · Cited by 1
“YOLO [56] is designed as the pioneering framework for one-stage detection, offering real-time processing capabilities by directly predicting bounding boxes and class probabilities from the entire image in a single evaluation.”
CrystalTracer: Automated Image Analysis of In Situ AFM Data Uncovers the Kinetics of Crystal Growth
Oluoch, L. S., Tao, J., & Clancy, P. (2026) · Crystal Growth & Design · doi:10.1021/acs.cgd.6c00581
Bus-Mounted Vision Sensing for Traffic Object Detection: BFTD and a Local–Global Attention Framework
Gao, W., & Zou, N. (2026) · Sensors · doi:10.3390/s26155001 · Cited by 1
“Classical CNN-based detectors such as Faster R-CNN [3], RetinaNet [6], and the YOLO family [7–9] have achieved strong performance by combining effective feature extraction with efficient inference.”
Gong, Z., Ge, N., Cao, S., 田开熙, & Yu, S. P. (2026) · IEEE Transactions on Neural Systems and Rehabilitation Engineering · doi:10.1109/tnsre.2026.3720921
“Handkinematics were captured using an Intel RealSense L515 LiDAR camera, combining YOLOv5 detection with hardware-level depth sensing for precise 3D reconstruction [36], [37].”
XLiteYOLOv12: a lightweight framework for real-time hazardous object detection
Yue, A., Chen, P., & Yang, Y. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01942-9
“In 2015, Redmon [11] proposed the first single-stage object detector YOLOv1, and Liu [12] et al. proposed the second single-stage object detector SSD.”
Advancing Wildlife Detection with YOLOv8 and YOLOv9: A Deep Learning Approach
Majmudar, M., Shah, H., Parmar, D., et al. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-26905-8_6
Liu, T., Wen, Q., Yang, J., et al. (2026) · Academic Radiology · doi:10.1016/j.acra.2026.06.070
A Hardware–Software Co-Design Optimization Framework for Edge-Oriented DETR-Based Object Detection
Zeng, X., Zhao, Q., Yu, N., & Hu, W. (2026) · IEEE Conference on Big Data Security on Cloud (BigDataSecurity) · doi:10.1109/bigdatasecurity69476.2026.00035
“To reduce computational overhead and achieve real-time performance, Redmon et al. proposed YOLO [16], a one-stage detector.”
Traffic-Burst-Resilient Queue-Adaptive Load Balancing for Edge-Assisted Autonomous Vehicles
Park, H., Lee, H., Jang, Y.-J., & Choi, J.-W. (2026) · International Conference on Ubiquitous and Future Networks · doi:10.1109/icufn69619.2026.11628616
“For parameter configuration, we utilized YOLO [21] to obtain the processing density, and BDD100K [22] datasets for video frame size.”
Pradhan, U., Ray, S., Aditya, M. N., & Anand, V. (2026) · Springer eBooks · doi:10.1007/978-3-032-26858-7_19
Liu, X., Hu, J., Jiang, Z., et al. (2026) · Sensors · doi:10.3390/s26165010
“In particular, one-stage detectors represented by the YOLO family have evolved from YOLO [4] to YOLOv7 [5], YOLOv11 [6], and YOLOv13 [7], and have been widely adopted in practical inspection tasks because of their favorable trade-off between detection accuracy and inference speed.”
Kanchanapradit, K., Noysena, K., & Lipikorn, R. (2026) · IEEE Access · doi:10.1109/access.2026.3721245
A Standardized Multi-Dimensional Evaluation Framework for Indoor Autonomous Service Robots
Chen, Q., Chen, Z., Maul, T., Zhou, D., Yang, Z., & Wen, J. Z. (2026) · International Conference on Artificial Intelligence, Robotics and Control (AIRC) · doi:10.1109/airc69745.2026.11631325
“Deep-learning-based object detection models, such as YOLO [19] and its subsequent variants, have significantly improved real-time perception performance on embedded platforms.”
Anomaly Detection System Design Based on Deep Learning for Tobacco Production Line
Song, M., Shi, Y., Zhu, L., et al. (2026) · doi:10.1109/iehns68708.2026.11606679
“The popular models include R-CNN [4], SSD [5], DETR [6], and YOLO [7].”
Detection of Crab Pots in Side-Scan Sonar Images Using Deep Learning Models
Abbasi, N. A., Bodine, C. S., Trembanis, A., et al. (2026) · doi:10.1109/oceans66983.2026.11617026
“YOLOv11 features an anchor-free detection head, transformer-enhanced backbone, and compound scaling techniques to optimize both performance and model complexity [32],[33].”
Zheng, J., Ju, B., & Sun, A. (2026) · Sensors · doi:10.3390/s26165030
“…convert spatiotemporal acoustic-field information into pseudo-color visual images and use target-detection networks such as YOLO to extract global spatial texture features, thereby improving the extraction of nonlinear and variable-speed PIG trajectories under moderate background interference [37].”
Ocean Monitoring with Underwater Robotics: Detection and Counting of Sea Urchins
Ji, Y., Zhu, A., Yuan, T., Hobson, B., Staples, G., Campbell, M., Desjardins, A. E., & Lampe, L. (2026) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans66983.2026.11616782
“You Only Look Once (YOLO) is a fast, single-stage convolutional network that predicts bounding boxes and class probabilities for all objects in an image [5].”
Adaptive Fusion of Detection and Density Networks for Enhanced Crowd Counting and Movement Analysis
Arora, K. R., & Sinha, A. (2026) · doi:10.1109/gseact68539.2026.11620568
“With the advent of the YOLO Family of Models [3], real-time object detection [7] became commonplace with the use of single-stage architectures that allow bounding box and class predictions [5] to be made simultaneously.”
Dileep, C., Arulselvi, M., & Reddy, K. P. (2026) · doi:10.1109/gseact68539.2026.11620123
“According to Szegedy et al., CNN architectures significantly enhanced feature extraction capabilities. it remains difficult to optimize these deep learning models for edge deployment [9].”
Wang, G., & Liu, D. (2026) · doi:10.1109/aic65131.2026.11633599
“Convolutional neural network (CNN)-based object detectors fall into two categories: two-stage methods—including R-CNN [5], Fast R-CNN [6], Faster R-CNN [7], Mask R-CNN [8], and Cascade R-CNN [9]—and single-stage approaches such as RetinaNet [10], GFL [11], TOOD [12], and the YOLO series [13], [14].”
Bajaj, A., Garg, R., Thakur, K., et al. (2026) · doi:10.1109/gseact68539.2026.11620029
“As summarized in Table 1 , the system integrates multiple modules—ranging from detection and alignment to similarity and semantic comparison—along with their corresponding operations, strategies, and objectives[10].”
Kiran, K., & Kumar, T. S. (2026) · doi:10.1109/gseact68539.2026.11620006
“HRNet [7] was proposed as the high resolution backbone network with parallel multi-resolution branches throughout the network and repeated information exchange between the branches.”
Detecting Driver Fatigue with Machine Learning: A Behavioral Approach
Skaf, Z., Wata, E. K., Sahilu, A. A., et al. (2026) · Advances in Science, Technology & Innovation/Advances in science, technology & innovation · doi:10.1007/978-3-032-24784-1_29
Analysis of event-based camera signals for Safety integration to IEC61499 *
Ladune, T., Colas, M., Philippot, A., Riera, B., Gellot, F., & Hampikian, P. (2026) · International Conference on Control, Decision and Information Technologies (CoDIT) · doi:10.1109/codit70676.2026.11631249
“The third group of approach such as presented in [12, 13, 14] consists of using neural network which can process sparse data.”
Zhao, P., Chen, W., & An, J. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01936-7
“One-stage detectors such as YOLO [14–16], SSD [17], and RT-DETR [18] directly predict object classes and locations, offering faster inference.”
High-Precision Orange Leaf Disease Detection Using Edge-Optimized Lightweight YOLO nano models
Bouallegui, M. A., Saidi, I., Abadi, A., et al. (2026) · International Conference on Control, Decision and Information Technologies (CoDIT) · doi:10.1109/codit70676.2026.11631220
“Among modern object detection models, the You Only Look Once (YOLO) family stands out due to its single-stage detection paradigm [1], [2], which combines high detection accuracy with fast inference speeds.”
Improved Real-Time Object Detection Using a YOLOv8-Based Ensemble Framework
Vashistha, P., Kayal, M., & Das, B. (2026) · IEEE Global Symposium on Emerging and Communication Technologies (GSEACT) · doi:10.1109/gseact68539.2026.11620141
“The YOLO model was one of the most significant additions to inference speed by re-formulating the task of object identification into a single regression problem [4].”
Kostarev, P., & Bartel, A. (2026) · International Conference on Control, Decision and Information Technologies (CoDIT) · doi:10.1109/codit70676.2026.11630706
“Each bounding box has a confidence score and class label [17].”
Chen, X., & Cheng, K. (2026) · International Journal of Computer Integrated Manufacturing · doi:10.1080/0951192x.2026.2714117 · Cited by 3
Advanced YOLO-Based Banknote Detection
Mirindi, F., Sinkhonde, D., & Mirindi, D. (2026) · IEEE International Conference on Computer Vision and Deep Learning (DLCV) · doi:10.1109/dlcv69906.2026.11635378
“Definitions appear in Table I. Anchor-based YOLO variants predict bounding boxes as offsets from predefined anchor priors [1], [14]; anchor-free architectures instead regress box centers and dimensions directly, which reduces sensitivity to hyperparameter choices.”
A Lightweight Model for PCB Surface Defect Detection
Yang, P.-C., Lee, W., & Wei, H. (2026) · Electronics · doi:10.3390/electronics15163500
A Lightweight In-Cabin Driver Object Detection Method Based on Improved YOLOv12
Han, K., Cheng, X., Li, D., et al. (2026) · doi:10.1109/dlcv69906.2026.11635544
“In contrast, YOLO-based methods [9] perform localization and classification in a more direct manner, making them more favorable for deployment because of their simpler structure and faster speed.”
An Overview of the Evolution of the YOLO Algorithm
Mirindi, F., Mirindi, D., & Sinkhonde, D. (2026) · IEEE International Conference on Computer Vision and Deep Learning (DLCV) · doi:10.1109/dlcv69906.2026.11635741
“Instead of first generating candidate regions and then classifying each region, YOLOv1 predicts bounding boxes, object confidence scores, and class probabilities directly from the full image in one network pass [1].”
InfluentialLv, T., Jiang, H., & Liu, Y. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3719466
“…fusion performance; in degradation scenarios, CLIP-IQA [46], MUSIQ [47], Q abf , and EN are adopted to assess degradation removal and information fusion e ff ectiveness; simultaneously, YOLOv8 [48] is used for object detection to verify the practicality of the fusion images in downstream tasks.”
Kethineni, K. K., Tang, A., Mohanty, S. P., & Kougianos, E. (2026) · Electronics · doi:10.3390/electronics15163512
Parlapalli, R. D., & Deepthi, K. (2026) · doi:10.1109/gseact68539.2026.11620229
Hu, H., 常朝乐, Hou, Q., & Zhang, C. (2026) · Developments in the Built Environment · doi:10.1016/j.dibe.2026.101003
Reddy, K. N., Swarna, B., Sathwik, G., Manikanth, B., Madhuri, G., & Gobbilla, U. R. (2026) · IEEE Global Symposium on Emerging and Communication Technologies (GSEACT) · doi:10.1109/gseact68539.2026.11620115
Consistencydet: few-step self-consistent box denoising for efficient object detection
Jiang, L., Wang, Z., Wang, C., Li, M., & Leng, J. (2026) · The Visual Computer · doi:10.1007/s00371-026-04674-w
“Early detectors relied on sliding windows or region proposals [7, 8]; anchor-based methods improved scale and aspect-ratio coverage [9, 10]; and recent query/proposal-based detectors, including DETR and Sparse R-CNN, recast detection as set prediction with learnable queries or proposals [11, 12].”
CEBW-YOLO11: A YOLO11-Based Model for Lithium Battery Electrode Sheet Defect Detection
Duan, G., Mi, Y., & Zhang, E. (2026) · Applied Sciences · doi:10.3390/app16167882
Wang, R., Tang, Q., Zhang, M., et al. (2026) · doi:10.1109/oceans66983.2026.11617048
“In contrast, single-stage detectors like YOLO [9] and SSD [10] directly predict target categories and positions on feature maps, achieving faster detection speed at the cost of relatively lower accuracy.”
Toshmatov, S. K., & Kosinskiy, V. V. (2026) · Zemleustrojstvo, kadastr i monitoring zemel' (Land management, cadastre and land monitoring) · doi:10.33920/sel-04-2607-04
Mask R-CNN Based Attention Guided CNN Ordinal Regression for Automated Car Damage Severity Detection
Biswas, S., Basak, D., Das, S., Saha, S., & Banerjee, S. (2026) · IEEE Global Symposium on Emerging and Communication Technologies (GSEACT) · doi:10.1109/gseact68539.2026.11620498
“One-stage detectors such as YOLO emphasize computational efficiency and are often adopted in time-constrained inspection workflows [6].”
AI-Based Grid Binary Annotation for Automated Recognition of Exfoliated Monolayer MoS 2
Lin, Z., Zhang, C., Pan, J., et al. (2026) · IEEE International Conference on Nano/Micro Engineered and Molecular Systems · doi:10.1109/nems69704.2026.11633730
“As a single-stage detector, YOLOv11 is renowned for its superior speed-accuracy trade-off and lightweight architecture [9], [10].”
Shi, J., Hu, R., & Chen, D. (2026) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans66983.2026.11617135
“As a result, Convolutional neural networks (CNNs), including real-time one-stage models such as the YOLO (You Only Look Once) series, have become the standard methodology [7].”
YOLO-Porpoise: A Cross-Region Robust Click Detection Model Based on YOLOv12
Gao, D., Jiang, Y., Lü, L., Du, L., Lü, Z., Yang, C., & Liu, Z. (2026) · MTS/IEEE OCEANS Conference (OCEANS) · doi:10.1109/oceans66983.2026.11617162
CCAN-YOLO: An Optimization of Aerial Small Object Detection Algorithm Based on YOLOv11
Zhang, D., Feng, G., & Han, R. (2026) · doi:10.1109/dlcv69906.2026.11635443
“One-stage detection algorithms such as SSD [1] and YOLO [2] series achieve fast inference speed relying on their concise network structure design and are suitable for real-time scenarios.”
Xu, S., Liu, X., Ren, Y., et al. (2026) · doi:10.1109/iehns68708.2026.11607448
“To achieve real-time performance, Redmon et al. [11] proposed YOLO (You Only Look Once), reformulating object detection as a single regression problem and enabling end-to-end detection at high speed.”
PWI-YOLO: An Improved YOLOv8 Model for Identifying Plateau Wetlands from Sentinel-2 Imagery
Zhang, Y., Zou, C., Wei, T., et al. (2026) · doi:10.1109/dlcv69906.2026.11635624
“The You Only Look Once (YOLO) algorithm partitions an entire image into a set of grids and independently classifies and predicts targets within each grid, effectively improving both detection accuracy and efficiency [8] , [9] .”
Knowledge amalgamation via images in the wild
Han, J., Fang, G., Guo, P., Song, J., Feng, Z., Wang, Y., & Song, M. (2026) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2026.104886
Ji, J., Wang, Y., Zhang, Y., Zhu, X., Guo, K., Sun, X., Chen, Q., & Niu, B. (2026) · Biology · doi:10.3390/biology15161344
“Convo-https lutional neural network (CNN)-based detectors and transformer-based object detection frameworks have demonstrated promising performance in camera-trap image processing, species recognition, and biodiversity monitoring [19–22].”
Attention-driven YOLOv11n–DeiT model for enhanced detection of tomato leaf diseases
Alsakar, Y. M., Sakr, N. A., & Elmogy, M. M. (2026) · Scientific Reports · doi:10.1038/s41598-026-64897-8
“Algorithms for real-time object detection, such as the YOLO family, have become fundamental in various tasks of computer vision because of their ability to perform accurate and fast detection in a single forward pass 17,56,57 .”
Huang, S., Zhang, Y., Wang, Y., et al. (2026) · Multimedia Systems · doi:10.1007/s00530-026-02599-x
Personal protective equipment detection method based on knowledge distillation
Zhou, C., Zhang, R., Fu, Y., et al. (2026) · Multimedia Systems · doi:10.1007/s00530-026-02574-6
“Early YOLO models already demonstrated significant speed advantages, with the fast version of YOLO [5] achieving 155 frames per second (FPS) and a mAP of 52.7% on the VOC07 dataset, while the enhanced version of YOLO reached 45 FPS with mAP of 63.4% on the same dataset [16].”
InfluentialDeep Learning Based Classification of Sedimentary Structures in Drill Core Images Using YOLO11
Muhammad, S. I., Kasim, S. A., Tijjani, I., et al. (2026) · SPE Nigeria Annual International Conference and Exhibition · doi:10.2118/234819-ms
Muhammad, S. I., Makwashi, N., Tijjani, I., et al. (2026) · SPE Nigeria Annual International Conference and Exhibition · doi:10.2118/234818-ms
Abdelaziz, A., Abdelwahab, H., & Ali, E. (2026) · Innovative Infrastructure Solutions · doi:10.1007/s41062-026-02873-x
Yuan, H., Wang, X., Liu, L., et al. (2026) · Journal of Supercomputing · doi:10.1007/s11227-026-08759-0
Yan, X., Tao, H., Xiong, B., et al. (2026) · Electronics · doi:10.3390/electronics15163547
Adaptive Spatial–Frequency Information Fusion for SAR Ship Detection
Xiao, Z., Zheng, X., Guan, D., et al. (2026) · Remote Sensing · doi:10.3390/rs18162687
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.