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 48,401–48,500 of 51,398
Tsan, Y., Wiryasaputra, R., Hsieh, Y.-J., Zhang, Q.-X., Liu, H. T., & Yang, C. (2026) · Diagnostics · doi:10.3390/diagnostics16091346
“Among existing approaches, the You Only Look Once (YOLO) family is particularly suitable for real-time applications due to its high inference speed [12,13].”
High-Accuracy Positioning Methodology for End Faces of Solid Propellant Cylindrical Grains
Wang, B., Cai, Y., Xu, J., et al. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11486972
Adaptive Hyperparameter Tuning for Deep Learning Based Object Detection in Side Scan Sonar Imagery
Besai, S. K., & Kumar, G. R. (2026) · IEEE International Conference on Intelligent Systems, Smart and Green Technologies (ICISSGT) · doi:10.1109/icissgt67741.2026.11483679
“The CNN-based computer vision models such as You Only Look Once (YOLO)[13], Single Shot Multi Box Detector (SSD) [14]and Faster R-CNN[15] rely heavily on hyperparameters like learning rate, batch size, confidence thresholds, and model-specific parameters.”
Thu, T. P. T., & Hong, T. N. (2026) · International Conference on Emerging Smart Computing and Informatics (ESCI) · doi:10.1109/esci68015.2026.11493380
Edge-Deployed Infrared-Based Occupancy Estimation for Privacy-Preserving Smart Environments
Chowdhury, D., & Chattopadhyay, M. (2026) · doi:10.1109/iccece69715.2026.11492835
“Recent deep-learning methods, like You-Only-Look-Once (YOLO) [8] and Faster R-CNN [9], have shown good results in low-occupancy scenarios, but conventional cameras capture identifiable details, raising privacy concerns [10].”
Li, Z. L., Ma, H., Chen, Z., et al. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487900
“The single-stage object detection framework known as YOLO (You Only Look Once) was initially pioneered by Redmon and Farhadi in 2017 [7].”
Kolla, N. K., & Vundavilli, P. R. (2026) · Modelling—International Open Access Journal of Modelling in Engineering Science · doi:10.3390/modelling7030083
Islas-Yañez, C. J., Hernández-Herrera, V., & Olivera, M. V. M. (2026) · Sensors · doi:10.3390/s26092785
A YOLOv8-Based System for Automated Archaeological Artifact Detection and Cataloguing
Ragab, Y. A., & Shamel, K. M. (2026) · International Youth Conference on Radio Electronics, Electrical and Power Engineering (REEPE) · doi:10.1109/reepe69046.2026.11481275
Improved RT-DETR algorithm for industrial defect detection
Wang, X., & Wu, J. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487639
“Specifically, one-stage algorithms approach the challenge of boundary localization by treating it as a regression task, so they do not need to generate candidate frames in advance, and they mainly contain the YOLO [4] series and SSD [5].”
Towards Object Counting by Efficient Scale Recognition Network
Zhai, W., Zhang, T., Zhang, T., Fan, Z., Gao, J., & Xing, X. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487391
A Wavelet-Based Efficient Defect Detection Model for Resource-Constrained Devices
Xiao, W., Wang, Y., Chen, Z., et al. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11486607
“In contrast, one-stage detectors, including the You Only Look Once (YOLO) series algorithms [3], [4] and RetinaNet [5], deliver high-speed detection at low computational cost.”
SSSD-YOLO: An Enhanced YOLO-Based Method for Steel Strip Surface Defect Detection
Yao, C., Xue, M., Wang, F., et al. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487898
“Current steel surface defect detection techniques can be categorized into three main approaches: image classification [4], [5], object detection [6]–[8], and image segmentation [9], [10].”
Prior-Knowledge-Guided and Model-Data-Driven Vibration Damper Defect Detection
Huang, Z., Fu, G., Chen, Y., et al. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487380
“Next, a YOLO-based [9] object detection network is constructed, consisting of three main components: the backbone, neck, and detection head.”
Yu, Q., Huang, R., & Li, G. (2026) · Systems · doi:10.3390/systems14050479
A Novel Rotated YOLOv8 for Mini LED Multi-Class Defect Detection
Zhou, L., Li, T., Chen, L., Zhang, K., Wu, Z., & Wang, W. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11486686
PER-YOLO: A Novel Vehicle Detection Model Based on YOLO11 for Snowy Weather Conditions
Kang, B. H., Chen, G., Zhou, L. H. T., et al. (2025) · China Automation Congress (CAC) · doi:10.1109/cac67268.2025.11487881
“In contrast, single-stage detectors like YOLO [2] are prevalent due to their efficiency, yet they still face low accuracy and missed detections in snowy scenes—plagued by snow occlusion and small-target ambiguity.”
Zhao, K., Wang, Y., Zhou, Z., Zhou, R., Zhang, X., Chen, X., Ran, D., Zhang, F., & Cao, L. (2026) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2026.3688483 · Cited by 1
“In computer vision, deep neural network (DNN) models such as ResNet [15] and YOLO [16] have established new standards for high-resolution analytics, which are essential for critical missions such as nat-ural disaster monitoring, precision agriculture [17].”
Application of Intelligent Aerial Robots to the Inspection and Maintenance of Electrical Power Lines
Ollero, A., Suárez, A., Marredo, J. M., Cioffi, G., Pěnička, R., Vasiljević, G., Hoang, V. D., Marolla, M., Xing, J., Saska, M., et al. (2024) · Emerald Group Publishing Limited eBooks · doi:10.1561/978-1-63828-283-920251009 · Cited by 1
SketchVision: Bandwidth-Efficient Edge-Cloud Video Object Detection via Compressed Sketches
Wang, H., Li, Q., Jiang, Z., Bao, Z., Jiang, Y., Yuan, Z., & Muntean, G. (2026) · IEEE Transactions on Broadcasting · doi:10.1109/tbc.2026.3680007
“In a typical real-time video object detection system, an edge device ( e.g., a camera) streams partial video to a cloud server, which runs object detection models ( e.g., YOLO [9]) to process user queries.”
AI for Corrosion Detection in Oil and Gas Refineries
Valavanis, L., Sarigiannis, G., Karydis, A., et al. (2024) · Emerald Group Publishing Limited eBooks · doi:10.1561/978-1-63828-283-920251004
IWAMOTO, T., Tanaka, S., & KAMEDA, Y. (2026) · Journal of the Japan society of photogrammetry and remote sensing · doi:10.4287/jsprs.65.83
Yu, W., Guo, X., Liu, X. Z., Liu, Y., Zheng, H., Huang, Y., & Yuan, Y. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01995 · Cited by 2
“Given a 2D X-ray image, regression-based methods aim to directly predict the camera pose using neural networks [6, 13, 18, 28], which is similar to traditional computer vision such as object detection [35, 43, 44] and 6-DoF object pose estimation [9, 22, 27, 42].”
Precision viticulture with AI: Comparing YOLOv5 models for grapevine fanleaf virus diagnosis
Şin, B., & Öztürk, L. (2026) · Tropical Plant Pathology · doi:10.1007/s40858-026-00803-x
YOLO-Based Real-Time Traffic Accident Detection for Smart Cities
Sunitha, G., Basi Reddy, A., Bhuvana, G., et al. (2026) · doi:10.1109/ccic68129.2026.11485997
“Redmon et al. [3] proposed a revolutionizing architecture termed You Only Look Once, which reformulated object detection as a regression problem, processing entire images in one pass.”
Vijay, M., Kota, S. H., Daka, S. N. P., et al. (2026) · International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET) · doi:10.1109/wispnet69615.2026.11489435
“The YOLO family [1] is a standard for real-time applications.”
Sadeq, M. A., Abduljabbar, A. K., Fayyadh, A. H., Aldulaimi, H. S., Mashhadany, Y. A., & Algburi, S. S. (2026) · International Conference on Electrical/Electronics, Robotics, Artificial Intelligence, and Informatics (ICERAI) · doi:10.1109/icerai69511.2026.11494538
“Accordingly, we propose an automated system that leverages computer vision and machine learning to detect uniforms in real time, using YOLO (You Only Look Once) [1] object detection techniques combined with face recognition methods to support real-time monitoring of uniform compliance.”
Natij, Y., Karch, H. E., Belaaribi, A., Lanaya, O., Atik, M. E., Maafiri, A., & Mezouari, A. (2026) · Frontiers in Plant Science · doi:10.3389/fpls.2026.1766704
Noise-aware Maritime Object Detection in UAV Imagery Using Joint Denoising-Detection Learning
Kaviyarasu, A., Nivashini, V., Rajesh, G., & Akther, M. R. (2026) · International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET) · doi:10.1109/wispnet69615.2026.11489438
“YOLO-based detectors are widely adopted in maritime UAV scenarios due to their real-time inference capability and efficient single-stage architecture [2].”
A Hierarchical Selection of Instance Segmentation Models for Wind Turbine Blade Defect Detection
Carballal, J. P., Maestrone, J. P., Olivera, M., et al. (2026) · IEEE International Conference on Industrial Technology (ICIT) · doi:10.1109/icit64854.2026.11491076
“One-stage object detectors, such as YOLO [4] and SSD [5], have been widely adopted for their balance of speed and accuracy.”
Development of Collision Alert System Using a Single Wide Angle Fovea Sensor
Shimizu, S., & Otsuka, K. (2026) · IEEJ Journal of Industry Applications · doi:10.1541/ieejjia.20250633
Feng, W. A. N. G., Li, X., Shan, X., Huang, H., Jiao, C., & JIANG, J. (2026) · Journal of Advanced Mechanical Design Systems and Manufacturing · doi:10.1299/jamdsm.2026jamdsm0013
Semi-Supervised Object Detection Based on Cross-Class and Cross-Model Non-Maximum Suppression
Yao, J., & Zhou, J. (2025) · International Conference on Big Data, Artificial Intelligence and Risk Management (ICBAR) · doi:10.1145/3800227.3800278
“Although these sequential filtering steps help ensure the quality of pseudo labels to some extent, they are not suitable for single-stage detectors that prioritize inference speed, such as YOLO[5].”
Su, Q., Xu, S., Wan, K., et al. (2026) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-6553-5_17
Human Fall Detection with Infrared Imaging: A Comparison of Graph Convolutional Networks and YOLO
Perliński, K., Faltyński, A., & Świetlicka, A. (2026) · Sensors · doi:10.3390/s26092794
“YOLO [4] is an image detection model that processes the entire image in a single pass, allowing it to simultaneously detect and localize objects with exceptional speed.”
Gobakken, H. R., Hoseini, M., Hoffmann, S., et al. (2026) · International Journal of Forest Engineering · doi:10.1080/14942119.2026.2656962
CaPTQ: Calibration Data Selection for Visual Services Based on Post-Training Quantization
Zhang, J., Yang, P., Ding, C., Raza, S., Zhao, Q., Yuan, P., & Wang, S. (2026) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2026.3689186
“Thus, this approach provides high-quality visual services, such as image classification [7], [8], [9], object detection [10], [11], [12] and tracking [13], [14], with lower resource consumption and faster response times.”
EdgeSecure-AI: A Multimodal Edge-Intelligent Threat Detection System for Smart CPS Surveillance
Raghul, S., Rohit, R., & Priya, S. (2026) · Contemporary Computing Innovations Conference (CCIC) · doi:10.1109/ccic68129.2026.11485972 · Cited by 1
Wang, R., Gupta, P., Ard, T., et al. (2026) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3626213
“YOLO v8 [30] and the Meta SAM model [31] are used to perform mixed reality tra ffi c image processing tasks.”
Beyond RGB: Adaptive Parallel Processing for RAW Object Detection
Gamrian, S., Barel, H., Li, F., Yoshimura, M., & Iso, D. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00526 · Cited by 4
Real-Time Seat Vacancy Detection in Public Transport Using YOLOv5-Based Deep Learning
Elakkiya, S., Amuthanantham, K., & Anandhi, S. (2026) · Journal of Soft Computing Paradigm · doi:10.36548/jscp.2026.2.002
ML AND RAG-BASED INTELLIGENT SYSTEM FOR YOGA POSE RECOGNITION AND CORRECTIVE GUIDANCE
Barapatre, H., Malgunde, P., Pratap, A., & Shaikh, R. (2026) · International Journal of Engineering Technologies and Management Research · doi:10.29121/ijetmr.v13.i4.2026.1768
MusicMamba: A Mamba-Based Automatic Singing Transcription Model
Chen, M., Li, T., He, Q., et al. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-8256-3_6
Marcos-Recio, S., Barrero-Bueno, A., Rossi-Labianca, L., Gil-González, A., Cardona-Mendoza, A., & Perdomo-Lara, S. (2026) · Applied Sciences · doi:10.3390/app16094464
Leveraging artificial intelligence to enhance marine biosecurity
Campbell, M. L., Le, C. T. U., Abu-Khalaf, J., & Sherman, C. D. H. (2026) · BioScience · doi:10.1093/biosci/biag049 · Cited by 2
Badminton Highlights Extraction in Live Streaming Using Deep Learning
Akila, K., Prasanna Raj, K. J., Nithish Kumar, S., & Manoj, A. (2026) · International Conference on Machine Learning and Autonomous Systems (ICMLAS) · doi:10.1109/icmlas67792.2026.11483821
“Convolutional Neural Networks (CNNs) and one-stage detectors such as YOLO are widely adopted for real-time object detection because of their favorable balance between accuracy and inference speed [1].”
Li, Z. (2026) · Journal of Computer Science and Artificial Intelligence · doi:10.54097/j3zk3g61
Improving the Performance of Fire Detectors Using Generative AI
Cho, W.-H., Kwon, O.-S., Kim, H.-Y., et al. (2026) · Fire science and engineering · doi:10.7731/kifse.0e1ed910
Huang, J., & Cai, Y. (2026) · International Conference on Big Data, Communication Technology and Computer Applications (BDCTA) · doi:10.1145/3801359.3801386
“In contrast, one-stage detectors represented by the YOLO series[18] achieve a superior trade-off between detection speed and precision.”
Real Time Facial Emotion Analysis with Quantum-Inspired CNN
Rajeswarappa, G., Prathima, C., Charishma, R., Kumar, K. P., Praneetha, P., & Parashara, S. C. (2026) · Contemporary Computing Innovations Conference (CCIC) · doi:10.1109/ccic68129.2026.11486241
“…real-time inference and is suited to be deployed on traditional computational platforms Experimental results have shown that the quantum-inspired architecture attains comparable performance with traditional CNN models while significantly reducing computational[5] complexity and training overhead.”
NMR as a Video-(Game): Constructing Super-Resolution Cross-Peak Trajectories in Protein Spectroscopy
Hang, C. J., Kuznietsova, Y., Fillipov, M., & Pervushin, K. V. (2026) · Russian Journal of Bioorganic Chemistry · doi:10.1134/s1068162026601771
Zhao, Y.-Z., Li, X., Zhao, Y., et al. (2026) · Complex & Intelligent Systems · doi:10.1007/s40747-026-02334-z
“In contrast, one-stage detectors such as YOLO [15–17] and SSD [18] are more efficient in terms of speed and deployment, making them suitable for real-time applications.”
Xia, Z., Guo, Z., He, K., Wu, D., Liu, C., Fu, Z., & Ma, L. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3687552
“These models include Faster R-CNN [46], Mask R-CNN [61], DESTR [62], YOLO series [63] (YOLOv8, YOLOv9, YOLOv10, and YOLOv11), deformable DETR [64], DINO [65], MaxViT [66], Efficient D3 [67], TOOD [68], TA-YOLO [69], DINO-v3-Small [70], RT-DETR [71], ESOD [72], HAFNet [73], YOLO26 [74], and RF-DETR…”
Smart Discipline Enforcement System for Educational Campuses
Sahoo, L., Mathumitha, R., & Mohana, M. (2026) · doi:10.1109/ccic68129.2026.11486021
Luan, W., Lu, S., Zheng, Y., Xu, W., Nie, L., Zhou, Z., & Liao, K. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02368 · Cited by 1
“Downstream Task Performance: Evaluated on SAM [18] and YOLO [32] in Tab.”
Zhang, Y., & Liu, C. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-45022-8.00004-8
NOGUCHI, H. (2025) · Journal of the Society of Biomechanisms · doi:10.3951/sobim.49.2_140
Liao, X., Wei, X., Leung, W. K., Yuen, Y. F., Yang, S., Wang, D., Zhou, Z., Xiao, X., Chen, Y., Huang, Y., et al. (2026) · IEEE Systems Man and Cybernetics Magazine · doi:10.1109/msmc.2025.3634086 · Cited by 1
A Detail-Enhanced YOLO Network for Steel Surface Defect Detection
Hao, Y., & Yuan, H. (2025) · International Conference on Big Data, Artificial Intelligence and Risk Management (ICBAR) · doi:10.1145/3800227.3800379
“The YOLO algorithm [2] is the most representative single-stage object detection method.”
Processing and Acquisition Traces in Visual Encoders: What Does CLIP Know About Your Camera?
Ramos, R., Stojnić, V., Kordopatis-Zilos, G., Nakashima, Y., Tolias, G., & García, N. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01584 · Cited by 3
“This core pipeline underpins various recognition tasks, from image classification [21, 35] and retrieval [34, 52] to object detection [55, 56] and segmentation [22, 33].”
Impact of Background Color on PET Preform Defect Detection Using a YOLO Based Deep Learning Model
Sribua, S., & Evans, W. K. (2025) · Artificial Intelligence and Cloud Computing Conference (AICCC) · doi:10.1145/3789982.3790009
“As a result, it is ideally suited for tasks like surveillance, automated monitoring, and other time-sensitive systems where immediate detection and classification of objects are crucial [13].”
A Lightweight Target Detection Algorithm for Remote Sensing Images
Zhou, L., & Chen, G. (2026) · International Conference on Big Data, Communication Technology and Computer Applications (BDCTA) · doi:10.1145/3801359.3801425
“One - stage algorithms like Centernet [4], SSD [5] and the YOLO series [6-7] directly predict target categories and locations by extracting features from the network.”
Wang, Z., Li, W., & Sun, K. (2026) · Remote Sensing · doi:10.3390/rs18091422
“Advances in deep learning have transitioned object detection algorithms from two-stage paradigms, like the R-CNN series [12], to one-stage detectors like SSD [13] and the YOLO family [14–22].”
Accurate 3D Object Detection with Minimal Labels: A CAM-Guided Weak Supervision Strategy
Saram, E. M. P. J. D., & Meegama, R. G. N. (2026) · International Conference on Image Processing and Robotics (ICIPRoB) · doi:10.1109/iciprob69625.2026.11497834
“To further accelerate object detection, single-stage detectors such as YOLO [5] and SSD [6] are introduced.”
Himabindu, B. A., Jyothi, G. N., Nandini, M. N. M., Sujatha, C., Manisha, N., & Subramanyam, M. (2026) · International Conference on Machine Learning and Autonomous Systems (ICMLAS) · doi:10.1109/icmlas67792.2026.11484018 · Cited by 1
“A comparison between these two models allows the proposed approach to give a clue on the performance of current YOLO models in medical imaging problems with microscopic lesions or stones[10].”
Esquivel-Félix, R., Moreno-Lucio, M., Castañeda-Miranda, C. L., et al. (2026) · Algorithms · doi:10.3390/a19050360
A Dataset of Ground-Captured Plant Images for Disease Detection and Categorization for Crop
A, N., M., P., Viswanath, G., et al. (2026) · 2026 IEEE International Conference on Emerging Computing and Intelligent Technologies (ICoECIT) · doi:10.1109/icoecit68303.2026.11497403
“Redmon et al. [19] made YOLO that is a real-time object recognition framework that could quickly find a leaf which is damaged.”
Performance Evaluation of Person Detection Using Scanless 3D Time of Flight Sensor
Okawa, H., Kawamata, M., Ito, T., et al. (2026) · International Conference on Image Processing and Robotics (ICIPRoB) · doi:10.1109/iciprob69625.2026.11497810
“Redmon et al. [4] proposes a person detection method using the You Only Look Once (YOLO) algorithm [5].”
Zhang, C.-L., Zhang, J.-M., Liu, H.-L., Qin, S.-Y., & Wang, J.-Q. (2026) · Nuclear Science and Techniques · doi:10.1007/s41365-026-01919-0
“Inspired by the You Only Look Once (YOLO) object detection algorithm [29], we proposed a novel full-energy peak region segmentation model called YOLOSpecNN.”
Zhou, Z., Liu, X., & Zhang, D. (2026) · International Conference on Electronics Technology and Artificial Intelligence (ETAI) · doi:10.1109/etai68332.2026.11485106
“To address the above limitations, this paper proposes a unified framework based on the YOLOv11 architecture (derived from the YOLO series models [8], [9]), achieving high-speed detection and precise spatial mapping through the integration of geometric topology modeling.”
MedROV: Towards Real-Time Open-Vocabulary Detection Across Diverse Medical Imaging Modalities
Sheikh, T. T., Lahoud, J., Anwer, R. M., Khan, F. S., Khan, S., & Cholakkal, H. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00832 · Cited by 5
“YOLO-World is an open-vocabulary object detection framework that extends the YOLOv8 architecture [40] by integrating vision-language modeling to enable detection beyond a fixed set of categories.”
Huang, Y., Liu, Y., Zheng, H., Qin, X., Xie, X., & Liang, Y. (2026) · Research Square · doi:10.21203/rs.3.rs-9455460/v1 · Cited by 1
Object Tracking and Shooting Turret
T., K. A., Murukan, V., P, A., et al. (2026) · 2026 IEEE International Conference on Emerging Computing and Intelligent Technologies (ICoECIT) · doi:10.1109/icoecit68303.2026.11496998
“[8] combined detection and localization in YOLO, thus, real-time accuracy that was perfect for turret applications became possible.Deep learning object framework has been depicted below in figure 3.”
InfluentialLow-Light Aerial Object Detection via Enhanced YOLOv11 with Feature Fusion
Xu, J., Fu, G., Wang, Y., et al. (2026) · Lecture notes in mechanical engineering · doi:10.1007/978-981-95-3019-9_30
Daksha, S., Gopinath, P., Manikandan, K. G., et al. (2026) · International Conference on Electronic Systems and Intelligent Computing (ICESIC) · doi:10.1109/icesic67389.2026.11496480
“This approach reduces reliance on manual inspections and improves hive management efficiency, supporting sustainable apiculture [5].”
Deepfake-Aware Face Authentication for Edge Devices Using a Unified Raspberry Pi Pipeline
Ghosh, S., Manish, S., Patra, P. K., Pattanaik, S. R., & Nayak, S. (2026) · IEEE International Conference on AI Engineering and Innovations (AIEI) · doi:10.1109/aiei69164.2026.11496983
“Within the last ten years, the use of deep learning algorithms along with Convolutional Neural Networks (CNN) has significantly improved the performances of face detection and recognition, where detectors like YOLO perform close to human performance [4].”
Rajesh, G., Nivashini, V., Akther, M. R., et al. (2026) · doi:10.1109/icmlas67792.2026.11483947
“When combined with modern object detection models such as the YOLO (You Only Look Once) family, drones can efficiently detect boats, swimmers, and other maritime objects, enhancing situational awareness and reducing response time [7], [8], [13].”
Paramesh, G., & Arulkumaran, G. (2026) · International Conference on Electronic Systems and Intelligent Computing (ICESIC) · doi:10.1109/icesic67389.2026.11496535
“Crucially, these perception techniques often function independently of the vehicle’s decision-making modules, or they rely on end-to-end architectures that lack the rule-level transparency required for safety-critical control [7].”
Manoj, A., Gupta, D., Krishnan, S. S. H., et al. (2026) · doi:10.1109/aiei69164.2026.11497587
“In contrast, single-stage detectors from the You Only Look Once (YOLO) family provide an effective solution due to their end-to-end architecture and real-time inference capability [7].”
Maiti, S., Dutta, R., Yadav, D. K., et al. (2025) · IEEE International Conference on Smart Technologies for Power, Energy and Control (STPEC) · doi:10.1109/stpec66316.2025.11490789
“The single-stage detectors, such as YOLO [5] and the modified versions of it, have revolutionized the single shot computer vision tasks since 2015, due to the faster performance as compared to the RCNNs [6], [7].”
Imitating the Functionality of Image-to-Image Models Using a Single Example
Spingarn, N., & Michaeli, T. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00216 · Cited by 1
Mondal, A., Mundra, S., Dasgupta, A., et al. (2026) · IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv61042.2026.00627
“In [5], the authors show that fine-tuning object detection models (Mask R-CNN [20], RetinaNet [30], SSD [31], YOLO [41]) on a closely related domain helps prevent overfitting and improves performance across tasks.”
Renedo-Alonso, I., Sánchez-Margallo, J. A., Arana-Arexolaleiba, N., & Elguea-Aguinaco, Í. (2026) · Sensors · doi:10.3390/s26092881
“Visual perception is implemented using a modular computer vision pipeline based on the YOLO (You Only Look Once) architecture [47], selected for its ability to perform real-time object detection with low computational overhead.”
Watanabe, S., & Premachandra, C. (2026) · International Conference on Image Processing and Robotics (ICIPRoB) · doi:10.1109/iciprob69625.2026.11497817
“Therefore, YOLO11, a YOLO variant that provides a favorable balance between detection accuracy and real-time performance, was selected and adopted in this study [15].”
BUDDY: Blind-Assistive Utility for Object Detection, Distance Estimation, and Voice Feedback
Singh, L., Gupta, S., Yashika, et al. (2026) · doi:10.1109/epdmc67535.2026.11496551
“YOLO variants, in particular, are favored due to their speed and efficiency for real-time applications compared to two-stage detectors like Faster R-CNN [12], [23][27].”
Infrared low-contrast small target detection method for UAV aerial images
Zhao, X., Zhang, H., Li, C., Wang, K., & Zhang, Z. (2026) · Infrared Physics & Technology · doi:10.1016/j.infrared.2026.106614 · Cited by 1
Learning-Based Real-Time Emotion Analytics for Multi-Person Settings with Confidence Visualization
Janjanam, V., Thallam, D. M., Spoorthi, B. G. P., et al. (2026) · doi:10.1109/aiei69164.2026.11496915
“To address latency, single-stage detectors like YOLO [4] and SSD have been developed, integrating face localization and emotion classification into a unified network.”
Karahan, F. E., & Özcan, G. (2026) · Eskişehir Türk Dünyası Uygulama ve Araştırma Merkezi Bilişim Dergisi · doi:10.53608/estudambilisim.1907006
Liu, Y., Wei, X., Chen, M., Liu, G., Zhao, C., Zhang, M., Guo, Y., Zhao, P., & An, R. (2026) · IET Image Processing · doi:10.1049/ipr2.70382
VLMs Guided Interpretable Decision Making for Autonomous Driving *
Hu, X., Jing, T., Tian, R., & Ding, Z. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00437
“To capture local objects, existing methods [11, 38] utilize ROI (Region of Interest) features from object detection or localization networks such as FastRCNN [8] or YOLO [28].”
Digital Recognition System of Traditional Chinese Patterns Based on Variant YOLOv8 Model
An, Z., Bai, J., & Xu, S. (2026) · doi:10.1145/3803730.3803732
“Its full compatibility with integrated training, validation [6], and export modes underscores its operational versatility and robustness in production environments, making it suitable for the end-to-end development of digital heritage applications [1, 3, 4].”
Quang, N. H., Lee, H., & Kim, G. (2026) · Natural Hazards · doi:10.1007/s11069-026-08173-y
“Dozens of well-known models have been developed, namely in employing the convolutional neural networks (CNN) only, such as YOLO (You Only Look Once) (Redmon et al. 2016 ), ResNet (Residual Network), introduced by Microsoft in 2015, EfficientNetV2 (Tan and Le 2021 ), Mask R-CNN (He et al. 2017 ), U-Net (Ronneberger et…”
Detection of Strong Gravitational Lenses with a Lens–Topology Relation and Arc-assisted Annotations
Meng, J., Wu, J., He, Y., Jiang, B., & Zhang, Y. (2026) · The Astrophysical Journal Supplement Series · doi:10.3847/1538-4365/ae5812
Complex Layout Analysis Algorithm in Chinese Ancient Mathematical Books
Guo, H., Li, Y., Dong, J., Zhang, C., & Gong, M. (2026) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-026-00584-8 · Cited by 1
Helmet Detection in Automated Teller Machine Surveillance Using Machine Learning
Aththanayaka, U., Silva, T., & Premachandra, H. W. H. (2026) · International Conference on Image Processing and Robotics (ICIPRoB) · doi:10.1109/iciprob69625.2026.11497809
Real-Time Lane-Wise Traffic Monitoring Using A Deep Learning Approach
Kirubakaran, N., Vardhan, C. V., Ahamed, S. M., et al. (2026) · doi:10.1109/nqcomp68334.2026.11497747
“Multi object tracking (MOT) systems using Deep SORT[12] or Fair MOT [13] solutions considered the aspect of identity preservation through the integration of motion and appearance clues.”
Zhou, C., & Chen, Q. (2026) · Measurement · doi:10.1016/j.measurement.2026.121733 · Cited by 2
Automated Traffic Flow Optimization Using YOLO and AI Models
Sinojia, K., Yeolekar, A., Vyas, J., & Bhonsle, M. (2026) · International Journal of Scientific Research in Science Engineering and Technology · doi:10.32628/ijsrset2613274
Industrial Case Study: Cost-Efficient Use of Cloud Spot Instances for Distributed DNN Training
Klemetti, A., Rosti, J., Raatikainen, M., & Nurminen, J. K. (2025) · Conference on AI, Science, Engineering, and Technology (AIxSET) · doi:10.1109/aixset65682.2025.00030
“Especially, the capabilities of Deep Neural Networks (DNNs) have revolutionized various tasks, including image processing [1]–[3], natural language processing (NLP) [4]–[6], and graph analysis [7], [8].”
Chakraborty, D., Haldar, S., Begum, S., & Das, S. (2025) · International Conference on Emerging Trends in Electronic Devices and Computational Techniques (EDCT) · doi:10.1109/edct65302.2025.11496040
“Since its introduction in 2015 [9], the YOLO architecture has evolved through multiple versions [10]–[12], culminating in the recent YOLOv8 [13].”
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.