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 47,401–47,500 of 51,398
Temporal-Consistent Pseudo-Label Learning for Domain Adaptive Object Detection
Chen, F., Li, J., & Liu, Z. (2025) · International Conference on Algorithms, Computing and Artificial Intelligence (ACAI) · doi:10.1109/acai68217.2025.11406524
“Traditional deep learning–based object detectors[1][2] rely heavily on large-scale labeled datasets to generalize effectively.”
Smart Assistive System for Pothole Screening using Lightweight CNN with Alert Generation
Das, A., Dasgupta, M., Nandi, D., et al. (2025) · 2025 IEEE 2nd International Conference for Women in Computing (InCoWoCo) · doi:10.1109/incowoco68239.2025.11407246
“Deep learning has revolutionized object detection with models such as YOLO (You Only Look Once) [4] and Faster R-CNN [5], which can detect pothole along with localization, which is widely used in research.”
GAI-YOLOv8: A precision-oriented detection algorithm for identifying fish targets in sonar images
Yao, J., Tong, J., Xue, M., & Ma, J. (2026) · Fisheries Research · doi:10.1016/j.fishres.2026.107701 · Cited by 1
Continuous Real-Time Image Processing by Variable Optical Processor
Miura, S., Otake, M., Nambara, T., Kusaka, H., Kunai, Y., & Kashiwagi, M. (2026) · IEEE photonics journal · doi:10.1109/jphot.2026.3669062
“Variousimage processingtaskssuchasclassification[4],[5],[6],[7],object detection[8],[9],segmentation[10],[11],super-resolution[12], [13],andimagegeneration[14]havebeenperformedbyAIs basedonneuralnetworks.”
Vaishnavi, K., Sathvika, M., Srinitha, P., et al. (2025) · doi:10.1109/siscon66686.2025.11409206
Robotic Assistive Smart Touch Inspection of Offshore Oil and Gas Pipelines
Chen, Z., & Song, G. (2026) · doi:10.70149/27941158
Steel Surface Defect Detection Using an Enhanced YOLOv8s with PPA Mechanism and AFPN
Wang, D. V. (2026) · Journal of Computing and Electronic Information Management · doi:10.54097/wfhyes67
“In recent years, deep learning-based object detection algorithms, particularly the YOLO (You Only Look Once) [1] series, have dominated the field of industrial defect inspection due to their remarkable balance between precision and inference speed.”
Best, P., Dassow, A., Kershenbaum, A., Nguyen, T. D., Pogson, M., Maheshwari, A., & Marxer, R. (2026) · PeerJ · doi:10.7717/peerj.20655
“We trained a You Only Look Once (YOLO) neural network ( Redmon et al., 2016 ) to detect ‘cao’ vocalisations within spectrograms.”
YOLO-Based Object Detection for Accurate Localization and Enumeration of Tomato Blight Pustules
Kumar, D., Kumar, N., Sharma, M. K., et al. (2025) · doi:10.1109/medcom67532.2025.11405304
Mutual Driven Network for Ship Detection in Foggy Conditions
Hu, R., Lü, T., Lang, X., et al. (2025) · International Conference on Algorithms, Computing and Artificial Intelligence (ACAI) · doi:10.1109/acai68217.2025.11406574
“The one-stage method directly classifies and locates from an input image, represented by the YOLO [16]–[18].”
Real-Time Monitoring Method for Drilling Site Cuttings Based on YOLOv12-seg
Yue, A., Chen, P., & Yang, Y. (2025) · doi:10.1109/mlcipr68329.2025.11407401
“With the introduction of the classic two-stage object detection model R-CNN [4] and the iterative optimization of single-stage models such as the YOLO series [5], these models can now efficiently process real-time video streams from shale shakers.”
DCAM-YOLOv8: A Dual Channel Attention-Based Architecture for Efficient Real-Time Object Detection
Vyas, R., Chunni, R., Bilgaiyan, S., et al. (2025) · OITS International Conference on Information Technology (OCIT) · doi:10.1109/ocit66168.2025.11400491
“Specifically, the You Only Look Once (YOLO) family of frameworks has emerged as one of the preeminent solutions for real-time detection due to its unified architecture and high inference rates [1]–[3].”
Blumenau, M. I., Khabarova, O., Никитин, И. М., & Vorobev, D. (2026) · The Astrophysical Journal · doi:10.3847/1538-4357/ae3f97
HAMD-DETR: A Wind Turbine Defect Detection Method Integrating Multi-Scale Feature Perception
Tian, S., Zhang, P., & Liu, L. (2026) · Energies · doi:10.3390/en19051235
Quantifying particle morphology via lightweight object detection and generative data augmentation
Liu, Z., Zhu, W., Qu, T., et al. (2026) · Acta Geotechnica · doi:10.1007/s11440-026-02949-z
A novel real-time image defect detection network for photovoltaic modules
Song, R., Ding, S., Zhang, Q., Su, J., & Chen, C.-W. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01862-8
“Single-stage detection architecture satisfies real-time requirements [16].”
Ma, Y., Lv, K., Liu, L., & Karimi, H. R. (2026) · Journal of Artificial Intelligence and Soft Computing Research · doi:10.2478/jaiscr-2026-0011 · Cited by 1
“The classical single-stage algorithms include OverFeat [16], Single Shot multibox-Detector (SSD) [17], You Only Look Once (YOLO) [18], etc.”
Towards Robust Multi-Object Pose Estimation Under Sparse and Noisy Data in Robotic Systems
Lee, H. (2025) · Conference on Cloud and Internet of Things (CIoT) · doi:10.1109/ciot67574.2025.11410147
“Deep learning (DL)-based object detection algorithms [3] have been widely adopted for object localization tasks due to their ability to generalize across various object appearances and conditions.”
Xing, Y., Ding, X., & Pan, H. (2026) · Signals · doi:10.3390/signals7020021
“This represents high-speed inference compared to state-of-the-art baselines while maintaining superior accuracy [31].”
KGCS: Zero-Annotation Expert-Knowledge Injection for Object Detection in Aerial Images
Hu, W., Hu, S., Ma, F., et al. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3670221
Research on Lightweight Pedestrian Detection Network Based on Improved YOLOv8
Dai, X., & Wu, H. (2025) · doi:10.1109/bdamea68159.2025.11406247
“Currently, the YOLO series of algorithms are extensively used in real-time object detection due to their superior speed-2025 accuracy balance [10-11].”
Real-Time Human Tracking and Predictive Danger Zone Forecasting for Public Safety
Govindarajan, K., & Rammohan, S. R. (2026) · International Conference on Computational Intelligence in Data Science (ICCIDS) · doi:10.1109/iccids69108.2026.11407654
“…Artificial Intelligence, especially related to deep learning based human detection algorithms such as YOLO variants, MobileNet-SSD, (Single Shot Multi-Box Detector), NanoDet [2], [3], have enabled more robust and high recognition accuracy of individuals in challenging environments of Indian cities.”
Dynamic Input-Aware Accuracy Control Hardware Architecture for Sobel Edge Detection
Harshavardhan, R., Goyal, K., & Rao, M. (2025) · IEEE International Conference on Emerging Electronics (ICEE) · doi:10.1109/icee67165.2025.11409859
“The proposed Sobel hardware is integrated with YOLOv5 [13] for targeted edge detection in Region-of-Interests (ROIs) via bounding boxes.”
Wang, J., Bie, H., Zhao, J., & Zhi, Y. (2026) · Scientific Reports · doi:10.1038/s41598-026-40379-9 · Cited by 3
Zhang, J., Li, D., & Luo, J. (2025) · doi:10.1109/ipmml68499.2025.11407407
“One-stage One-stage methods, first proposed by Joseph et al.[14] in 2015, are renowned for their efficient inference speed but generally suffer from lower accuracy and poor performance in detecting small objects.”
Real-Time Driver Drowsiness Detection System Using Ai-Enhanced Vision
Vanitha, U., Amirtha, R., Deepakarthika, M., Gokulmathi, P., & Soundari, D. V. (2025) · IEEE International Conference on Innovations in Engineering and Next-Generation Technologies for Sustainability (ICINVENTS) · doi:10.1109/icinvents64613.2025.11401404
ASL-YOLO: A Lightweight and Efficient YOLOv11 Model for Rail Surface Defect Detection
Lin, Y., & Mou, L. (2025) · International Conference on Signal Processing, Computer Networks and Communications (SPCNC) · doi:10.1109/spcnc68200.2025.11406249
“In contrast, one-stage approaches such as SSD [3] and the YOLO series [4] omit an explicit proposal stage and learn to output bounding boxes together with class scores in a single forward pass, which makes them easier to deploy and better suited for real-time applications.”
Open-Vocabulary vs Supervised Learning Methods for Post-Disaster Visual Scene Understanding
Michailidou, A., Angelidis, G., Argyriou, V., Sarigiannidis, P. G., & Papadopoulos, G. T. (2026) · International Conference in Electronic Engineering & Information Technology (EEITE) · doi:10.1109/eeite69609.2026.11679544
“For object detection in particular, the field has evolved from fast one-stage detectors [14] toward end-to-end set prediction with transformers [15], including designs targeting real-time operation [16], [17].”
Zhang, C. (2025) · International Conference on Computer, Internet of Things and Smart City (CIoTSC) · doi:10.1109/ciotsc67482.2025.11413161
“SGD was selected over adaptive optimizers like Adam because it performs well in large-scale image classification and detection tasks [1]; by introducing more noise into gradient updates, SGD helps reduce overfitting — a key advantage for crack detection, where the model must avoid learning false…”
Underwater Structural Damage Detection of Bridges Based on an Improved YOLOv11n Algorithm
Zhao, J. (2025) · International Conference on Computer, Internet of Things and Smart City (CIoTSC) · doi:10.1109/ciotsc67482.2025.11412998
“As a single-stage detector, the YOLO series performs end-to-end object detection rather than using multi-stage pipelines [16].”
Weld Defect Detection Algorithm Based on ADD-YOLO
Zhu, L., & Mou, L. (2025) · International Conference on Signal Processing, Computer Networks and Communications (SPCNC) · doi:10.1109/spcnc68200.2025.11406661
“In contrast, single-stage approaches, exemplified by the YOLO [2] family, employ a unified convolutional neural network to achieve end-to-end feature extraction, object classification, and localization, thereby markedly improving detection efficiency.”
TANG, Y., Liu, B., Su, Y., et al. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3669904
Zhao, W., Liu, Y., Wen, L., et al. (2026) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.104535
Survey of AI-Driven Adaptive Traffic Signal Detection Using Edge–IoT Architecture
Vishwakarma, L. (2026) · Interdisciplinary Journal of AI Machine Learning & Data Science · doi:10.66261/mdhcw538
Framework for Gym Posture Estimation Using MediaPipe and OpenCV
Patidar, K., & Zaveri, M. A. (2025) · IEEE International Conference on Recent Advances in Computing and Systems (REACS) · doi:10.1109/reacs67479.2025.11413297
“Unlike other generic pose recognition systems such as YOLO [3] based detection frameworks, this implementation is optimized for single-person exercise analysis with low computational requirements and high frame-rate efficiency.”
Redes Neurais Convolucionais Aplicadas à Odontologia: Revisão Sistemática da Literatura
Rodrigues, M., de Sousa, R. M., de Lima, B. V. A., & Santos, I. G. D. S. (2025) · doi:10.5753/codec.2025.39148
Osmerova, L., Kopecka, E. G., Opelka, J., Begall, S., Hrouzková, E., Bondar, A., Sehadová, H., Šumbera, R., & Gahurová, L. (2026) · Mammalian Biology · doi:10.1007/s42991-026-00569-4
Lee, H. W., & Byun, J. (2026) · Korean Journal of Applied Statistics · doi:10.5351/kjas.2026.39.1.065
Shelf Image Analysis Using YOLO11 and YOLOv12 for Product Recognition
Saminathan, R., & Bagyammal, T. (2026) · International Conference on Mobile Computing and Sustainable Informatics (ICMCSI) · doi:10.1109/icmcsi67283.2026.11412451
“In this study we propose the use of the latest models released for the YOLO [1] algorithm, including YOLO11 [3] and YOLOv12 [4] in small and medium variants on the Grozi-3.”
Deep Learning-Based YOLO Framework for Accurate Detection and Coordination of Autonomous Vehicles
Reddy, V. S., Rinku, D. R., Juluru, T. K., Chitti, S., Ramchandar Rao, P., & Ramadevi, R. (2026) · International Conference on Mobile Computing and Sustainable Informatics (ICMCSI) · doi:10.1109/icmcsi67283.2026.11412499 · Cited by 1
“The YOLO algorithm works mainly on three techniques, those are [14],[15]”
Farag, W. A. (2026) · Journal of Electrical Systems and Information Technology · doi:10.1186/s43067-025-00303-z
“Models like YOLO [34] (You Only Look Once) [35] and SSD (Single Shot Multibox Detector) [36] are widely used in autonomous driving for tasks such as pedestrian detection and vehicle recognition.”
Smart Retail Monitoring: An End-to-End Transformer-Based System for Shelf Intelligence
Anitha, G., Ryan, J. C., Shashini, V., & Reddy, G. V. N. (2025) · Global AI Summit - International Conference on Artificial Intelligence and Emerging Technology (AI Summit) · doi:10.1109/aisummit66170.2025.11411620
PFM-TurtleNet: Sea Turtle Species Identification Using a Parallel Fusion Module
Prasetya, H., Nguyen, D.-L., Pantouw, S. O., et al. (2026) · ELCVIA Electronic Letters on Computer Vision and Image Analysis · doi:10.5565/rev/elcvia.2297
“Among the (CNN) architectures, YOLO (You Only Look Once) stands out as a high-performance object detection framework characterized by its computational efficiency and real-time processing capabilities[10, 24].”
Wang, Q., Li, P., Li, Y., Zhang, L., Niu, B., Wang, F., Geng, X., & Zhou, G. (2026) · Remote Sensing · doi:10.3390/rs18050772
Improved YOLO11-Based Low-Light Vehicle and Pedestrian Detection Algorithm Using Image Enhancement
唐, 翱. (2026) · Journal of Sensor Technology and Application · doi:10.12677/jsta.2026.142030
Bangalisan, H. Y., Sneha, J. A., & Sivakumar, V. (2025) · doi:10.1109/wecon68556.2025.11414561
“These models were trained using Roboflow’s cloud-based training environment [35].”
Wang, D., Jiang, W., Shi, Z., et al. (2025) · doi:10.1109/ichbc68357.2025.11414586
“[10 ] The final output of the network is a three-dimensional tensor, which, after post-processing steps like confidence threshold filtering and non-maximum suppression (NMS), yields the final detection results.”
Wu, Z., Jiang, Y., Du, H., et al. (2025) · IEEE International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412705
“Since its initial introduction by Joseph Redmon in 2016 [13], YOLO technology has undergone multiple iterations, each bringing significant improvements in performance, speed, and accuracy.”
Cheng, M.-Y., Khitam, A. F. K., Hartono, R. A., & Chuang, T. (2026) · International Journal of Construction Management · doi:10.1080/15623599.2026.2630245 · Cited by 1
Forest Fire Object Detection Based on an Improved YOLOv5s
Li, Z., Xiang, S., Li, X., et al. (2025) · IEEE International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412879
“For this experiment, precision ( P ), recall ( R ), average precision ( AP, AveragePrecision ), mean average precision ( mAP, meanAveragePrecision ) [16]–[18], and detection speed FPS have been used as evaluation metrics.”
SPBNet-YOLO: Improving Small Object Detection in UAV Imagery Under Diverse Conditions
Tariq, M. N., Zhang, H., Khan, M. R., & Zhao, S. (2025) · International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412919
“YOLO [6], RetinaNet [7], CenterNet [8], SSD [5], Fast R-CNN [3], Faster R-CNN [4], and R-CNN [2] are examples of popular detection frameworks.”
Penta, A., Adwani, V., & Chopra, A. (2025) · IEEE International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412868
Real Time Object Detection using Yolo and Open CV
Pawar, S., Suma, K., Geethalaxmi, et al. (2025) · doi:10.1109/wecon68556.2025.11414583
“Redmon et al. [2] improved YOLO by making it as revolutionized object detection by making it as a single regression problem.”
A Resolution-Reconfigurable SAR-ADC for Computing-in-Memory to Break Power Bottleneck
Liu, T., Zhang, Q., Li, C., et al. (2025) · doi:10.1109/ichbc68357.2025.11414622
“The rapid development of Internet of Things (IoT), edge intelligence, and real-time perception applications has driven the migration of smart sensing tasks such as image classification and object detection [1] [2], from the cloud to near-sensor edge devices.”
Impact of Different Rain Removal Preprocessing Methods on Rainy Object Detection Performance
Dong, J., Su, D., He, S., Tang, H., Liu, L., & Xie, Z. (2025) · International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412946
“With the rapid development of deep learning technology, convolutional neural network models represented by Faster R-CNN and YOLO [4] emerged, which can automatically learn the features of input images, thus significantly improving the performance of target detection.”
Deep Neural Network Utilization in Wildfire Hotspot Classification For Drone-Obtained Imagery
de Jesus, L. D., Dantas, A., Shiguemori, E. H., & Vijaykumar, N. L. (2025) · IEEE Latin American GRSS & ISPRS Remote Sensing Conference (LAGIRS) · doi:10.1109/lagirs68367.2025.11414805
“However, the YOLO (You Only Look Once) family of models has gained prominence due to its balance between accuracy and inference speed, which is particularly relevant in real-time scenarios [16].”
Improved YOLOv8n-based Algorithm for Road Object Detection in Foggy Weather
Huang, W., Chen, Z.-Y., Zhang, D., et al. (2025) · 2025 IEEE 8th International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412915
GTWNet: A High-Accuracy and Lightweight Framework for Rice Pest Detection
Deng, S., Wu, G., & Zhang, W. (2025) · International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412638
“To address the complex demands of pest and disease detection in agricultural production, especially in scenarios requiring both speed and accuracy, YOLO [22] series models have become a research hotspot due to their efficient detection capabilities.”
Zhang, X. (2025) · International Conference on Electronic Information Engineering and Computer Communication (EIECC) · doi:10.1109/eiecc67963.2025.11409620
“Another type is the single-stage detection method represented by YOLO[1], which directly performs position regression.”
A Defect Detection System for Printed Circuit Boards Based on Enhanced HFB-YOLO Architecture
Hou, K. (2025) · doi:10.1109/ccnis69465.2025.00091
“In contrast, one-stage algorithms, such as the YOLO (You Only Look Once) series [9] and SSD [8], treat object detection as a single regression problem.”
Performance Evaluation of YOLOv11 for Ship Recognition in Complex Sea Surfaces
Xue, H., Chen, C. L. P., Shao, X., et al. (2025) · IEEE International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412463
“Redmon first proposed the YOLO algorithm v1 version [17] in 2016, reducing the target recognition problem to a regression problem.”
Li, C., Liu, J., & Qu, Z. (2026) · Signal Image and Video Processing · doi:10.1007/s11760-026-05204-w
LA-DETR: Local Attention and Multi-Scale Aggregation for Real-Time Steel Defect Detection
Fu, R., Tong, J., Zhao, Q., et al. (2025) · International Conference on Electronic Information Engineering and Computer Communication (EIECC) · doi:10.1109/eiecc67963.2025.11409533
“SSD [7] employs multi-scale feature maps for efficient detection, while the YOLO series [6], [12], [13] treats detection as a unified regression problem, achieving real-time performance through single-pass inference.”
Li, B., Peng, C., Zou, K., & Zhou, W. (2025) · International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412830
“The YOLO (You Only Look Once) [13] framework was first introduced by Redmon et al [14]. in 2016.”
Research on Flame and Smoke Detection Algorithms Based on Deep Learning
Wu, Z., & Li, Q. (2025) · International Conference on Electronic Information Engineering and Computer Communication (EIECC) · doi:10.1109/eiecc67963.2025.11409604
MTL-YOLO: Multi-Task Learning-Based YOLO for Detecting Bone Irregular and Discontinuity
Shang, S., Cao, M., & Zhao, H. (2025) · IEEE International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai67447.2025.11412477
“Furthermore, variations in the size and shape of bone discontinuity across different imaging modalities place high demands on the adaptability and feature extraction capabilities of detection method [5].”
Yin, L., & Wu, J. (2025) · International Conference on Electronic Information Engineering and Computer Communication (EIECC) · doi:10.1109/eiecc67963.2025.11409634
“YOLOv8n is a popular variant of the YOLO [9][10] architecture, well-suited for accurate defect detection.”
Zaznava in klasifikacija brezpilotnih letalnikov s kamero in radarjem
Andov, P., Smogavec, P., & Gleich, D. (2026) · doi:10.18690/um.feri.4.2026.11
Real-Time Fire Detection System for Critical Infrastructure with Cost-Sensitive Learning
Pravesh, R., & Sahana, B. C. (2026) · Journal of The Institution of Engineers (India) Series B · doi:10.1007/s40031-026-01299-4
“With the rise of CNNs, both single-stage (Single Shot MultiBox Detector [12], and RetinaNet [13], YOLO [14–19]) and two-stage (Faster R-CNN [20], Cas-cade R-CNN [21], Mask R-CNN [22]) detectors have been explored for real-time fire identification.”
de Haro, S., Cámara, J., González‐Férez, P., García, J. M., & Bernabé, G. (2026) · Algorithms · doi:10.3390/a19030200
A methodological scoping review for video analysis in hospitality research
Xu, T., Erdem, M., Abarbanel, B., & Kraus, S. W. (2026) · Journal of Hospitality and Tourism Insights · doi:10.1108/jhti-10-2025-1183 · Cited by 1
Ma, K., Zhang, Z., Zhang, J., & Huang, J. X. (2026) · Electronics · doi:10.3390/electronics15051100 · Cited by 2
Li, X., Yang, W., Wu, J., Wang, Z., Tian, Q., Ye, T., Chen, W., Yu, F. R., & Leung, V. C. M. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3671257 · Cited by 1
“YOLO v1 [20] pioneered real-time detection by treating it as a regression problem.”
Raj, A., & Prethija, G. (2026) · Machine Learning with Applications · doi:10.1016/j.mlwa.2026.100884
BenDan: Benchmarking DPU performance on FPGAs
Bao, H., Liu, X., Wang, X., Sadaqa, A., Zhu, Y., Tang, S., Chen, R., Ling, M., & da Silva, B. (2026) · Integration · doi:10.1016/j.vlsi.2026.102695
Qu, Z., Zang, J., Yue, Y., Li, L., Qu, N., Huo, G., & Bo, X. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131951
Recent Trends in Deep Neural Networks and Their Hardware Implementation for Biomedical Applications
Monga, A., Nehete, H., Dhull, S., Nisar, A., Yadav, S., & Kaushik, B. K. (2026) · Wiley eBooks · doi:10.1002/9781394336203.ch12
LightIR-YOLO: A Lightweight YOLO for Infrared Drone Detection
Niu, H., Zhang, J., Chen, W., et al. (2025) · doi:10.1109/icusac68314.2025.11414353
“Representative algorithms include the R-CNN series [1-2] , the YOLO series [3-9] , and the DETR series [10-13] .”
Hybrid CNN-Transformer Architecture with Lightweight Modules for Pavement Defect Detection
Zhou, G., Du, Z., & Zhou, Y. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00176
“Currently, mainstream meth-ods include single-stage object detection algorithms (such as YOLO [1], SSD [2], RetinaNet [3] and RT-DETR [4]) and two-stage (such as Faster R-CNN [5] and Mask R-CNN [6]).”
Habibi, L., Maïdi, M., Boubchir, L., et al. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401600
Drone jamming decision-making method based on dynamic Shapley value
Yan, J., Chen, P., He, Z., et al. (2025) · doi:10.1109/icusac68314.2025.11414340
“Third, the game theory method [5] analyzes participant cooperation, but it assumes a static environment and does not consider the contribution changes over time.”
Multi Palm Oil Tree Identification Using Geo-Referenced Deep-SORT for Unmanned Aerial Vehicle
Suharsono, A. D. P., Mardiyanto, R., & Purwanto, D. (2025) · International Conference of Science and Information Technology in Smart Administration (ICSINTESA) · doi:10.1109/icsintesa68165.2025.11413684
“It employs a Convolutional Neural Network (CNN) to directly predict bounding boxes and class probabilities for each object from the raw pixels of an input frame [15].”
Wang, F., & Geng, S. (2025) · International Conference on Advanced Computer Theory and Engineering · doi:10.1109/icacte66284.2025.11411977
“EfficientDet [12] advanced multi-scale fusion via BiFPN and compound scaling, while YOLO [13]–[17] revolutionized the field with its unified regression approach, with YOLOv8 now setting industry standards through anchor-free detection and optimized architecture, delivering unparalleled…”
Research on Small Target Detection Algorithm for UAV Aerial Images
Liu, M., & Li, X. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00050
“…progress: two-stage algorithms (e.g., R-CNN [1] and Faster R-CNN [2]) stand out for their superior performance, single-stage models (e.g., YOLO [3]–[6] and SSD [7]) excel in inference speed, and Transformer [8]-based architectures (e.g., Swin-Transformer [9]) have injected new vitality into…”
CoP: Chain of Perception for Referring 3D Instance Segmentation
Ma, Y., Ji, J., Qian, Z., Sun, X., & Ji, R. (2026) · International Journal of Computer Vision · doi:10.1007/s11263-025-02699-7
WD-YOLO: A Road Damage Detection Framework via EFWaveletPool and WDRGFPN
Liao, R., Jiang, T., Tang, S., et al. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00143
“Existing algorithms are divided into two-stage (e.g., R-CNN [2]) and single-stage (e.g., YOLO [3] series, SSD [4] series) categories: the former improves accuracy through region proposal and fine detection, while the latter ensures real-time performance with an end-to-end design.”
Mohammed, A. S. (2026) · doi:10.1109/elcon-cn69892.2026.11414018
“This is 1 for bounding box predictors that are not responsible for an object [17].”
Analysis of the performance of the detection of cassava leaf diseases using YOLO versioning
Sore, S. S., Bikienga, M., & Frédéric, O. T. (2025) · 2025 5th International Conference of Science and Information Technology in Smart Administration (ICSINTESA) · doi:10.1109/icsintesa68165.2025.11413807
“A study presented in [17] compared the performance of Faster R-CNN and YOLOv5.”
CO-YOLO: Collaborative Enhancement Network for Efficient Low-Altitude Object Detection
Zhao, Y., & Wu, X. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00063
“Cascade R-CNN [17] em-ployed a sequence of cascaded detectors to enhance detection quality and reduce overfitting.”
Smart Scale Food Analysis for Stage 1 Cancer Patients
Hs, S., M., A., R., C., et al. (2025) · 2025 First International Conference on Intelligent Computing and Systems at the Edge (ICEdge) · doi:10.1109/icedge67252.2025.11412410
“Various deep learning techniques for object detection have been proposed in recent years[2-4].”
A Lightweight Real-Time Framework for Skeleton-Based Action Recognition on Mobile Devices
Zhang, Q., Wang, W. L., Zhang, B., Zhang, Z., Que, X., & Wang, W. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401553
Mao, Q., Wei, F., Neary, R., Wang, C., Qin, H., Zhang, J., & Huber-Fliflet, N. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11402368
Real-Time Object Detection Technology for Complex Underwater Environments Using Improved YOLOv10s
Fang, Z., Pan, Z., & Ding, Z. (2025) · International Conference on Advanced Computer Theory and Engineering · doi:10.1109/icacte66284.2025.11412203
“With the continuous advancement of deep learning techniques [5], object detection algorithms such as YOLO [6], Single Shot MultiBox Detector (SSD) [7], and Faster-RCNN [8] have been extensively applied to various underwater image detection tasks [9], demonstrating promising performance in…”
Syed, F. H., Chatterjee, A., Chatterjee, M., et al. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401571
“Among the leading frameworks, YOLO (You Only Look Once), introduced by Redmon et al. in 2015 [15] and it’s state-of-the-art architecture YOLOv5 [11] has emerged as a dominant mode due to its balance of speed and accuracy.”
Efficient Detection of Tibetan Barley Growth Status via DINO-Guided Knowledge Distillation
Wang, Y. P., Guang, H., Bai, Y., et al. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00145
ReLAD: Retrieval-Augmented LAtent Diffusion for Complex Aerial Image Generation
Townsell, D., Chen, L., Xie, M., Pan, C., & Zhang, W. (2025) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata66926.2025.11401711
“First, we train an object detection model based on YOLO architecture [36] using the aerial dataset to detect areas of interest.”
Sahro, A., Budi, E. M., & Nadhira, V. (2025) · International Conference of Science and Information Technology in Smart Administration (ICSINTESA) · doi:10.1109/icsintesa68165.2025.11413818
Cui, C., Sun, Y., Li, L., et al. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00147
“Modern object detection is dominated by CNN-based architectures [7], [9]–[11], where multi-scale designs like the Feature Pyramid Network (FPN) [4] and its variants [5], [14] are standard for handling scale variance.”
LiteFuseNet:A Lightweight Edge-Aware Multi-Scale Detection for Mobile Robots
Li, W., Qi, J., Cui, Z., et al. (2025) · International Conference on Virtual Reality and Visualization · doi:10.1109/icvrv67992.2025.00030
“Two-stage detectors such as Faster R-CNN [3] generate region proposals followed by classification and bounding box regression, whereas one-stage detectors like YOLO [4] and SSD [5] directly predict object categories and bounding boxes from dense feature maps.”
Pan, H., & Guo, Q. (2026) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-026-01339-5
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.