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 46,401–46,500 of 51,399
Real-Time Gun and Person Detection for Smart Surveillance
Connan, J., Brown, D., Marais, M., & Goodall, L. (2025) · International Conference on Artificial Intelligence, Big Data, Computing and Data Communication Systems (icABCD) · doi:10.1145/3759023.3759128
“…monitoring video There is a greater body of research for older object detection models such as Faster Region-based Convolutional Neural Network (Faster R-CNN) [36], single-shot multi-box detector (SSD) [25] and early iterations of You Only Look Once (YOLO) [35] in the context of surveillance.”
Mobile Application for Object Height Measurement
Kumari, L. V. R., Yoshitha, E., Bhargav, V., et al. (2025) · Lecture notes in networks and systems · doi:10.1007/978-981-96-9196-8_9
梁, 秉. (2025) · Computer Science and Application · doi:10.12677/csa.2025.1510257
Application of YOLO in Indian Driving Conditions
Mohan, S. G., Raj, A. K., Nayana, S. A., et al. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-06662-6_6
PEAK: Generating High-Performance Schedules in MLIR
Tavakkoli, A. M., Joshi, S., Singh, S., Xu, Y., Sadayappan, P., & Hall, M. W. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-02436-7_13
A lightweight and multi-scale model for MLCC appearance defect detection
Zheng, J., Cao, X., Zhang, D., Takamasu, K., & Chen, M. (2025) · Optics & Laser Technology · doi:10.1016/j.optlastec.2025.114173
Zhao, X., Huang, Y., He, Q., Lu, X., & Li, J. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.134112
Lightweight Temporal Consistency for Grid-Based Obstacle Detection in Edge Devices
Kurkutlu, O., & Roohi, A. (2025) · IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C) · doi:10.1109/acsos-c66519.2025.00047
“Traditional object detection methods, such as Faster R-CNN [14], SSD [15], and YOLO [16], have demonstrated strong performance on general-purpose GPUs and CPUs.”
Dai, L., Yan, X., Xiong, J., Zhang, S., & Zhai, X. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-3736-5_16 · Cited by 1
Edge Feature-Enhanced Swin Transformer for Insect Image Classification
Li, Y., Zhang, L., & Sun, L. (2025) · Communications in computer and information science · doi:10.1007/978-981-95-3736-5_20 · Cited by 1
YOLO-OptiMob: A Pipeline for Optimizing YOLO11 Models for Edge Deployment
Bhat, A. M., Eeragond, A. M., Nandyal, J., et al. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-96-9971-1_21
Obj3Dify: Occlusion-Invariant 3D Reconstruction of Hand-Held Objects
Mulero-Pérez, D., Ruiz-Ponce, P., Benavent-Lledo, M., Ortiz-Perez, D., & Garcia-Rodriguez, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229385
Zhang, W., Wang, Z., Yuan, X.-Q., et al. (2025) · Signal, Image and Video Processing · doi:10.1007/s11760-025-04957-0
“The YOLO model proposed by Joseph et al. [10 ], as the first deep learning-based one-stage target detection, uses an end-to-end idea to achieve detection regression through a convolutional neural network, resulting in a substantial increase in detection speed, then Joseph designed YOLOv3 [11] with…”
Dyadic Human-Robot Interaction: Emerging Technologies, Challenges, and Opportunities
Adebayo, S., Dessing, J. C., & McLoone, S. (2025) · Transactions on computer systems and networks · doi:10.1007/978-981-96-9471-6_1 · Cited by 1
Indoor Obstacle Object Detection Algorithm Based on Improved YOLOv8n
Song, Z., & Lu, J. (2025) · International Symposium on Artificial Intelligence and Intelligent Manufacturing (AIIM) · doi:10.1109/aiim67611.2025.11232846
“The YOLO series of algorithms [8] proposed by Joseph Redmon et al. adopts an end-to-end architecture for real-time detection, with YOLOv3 improving small object detection accuracy to 28.2%, but still facing missed detections in dense scenes.”
InfluentialNguyen, T.-V.-A., Le, T.-H., Hoang, Q.-V., & Pham, N.-T. (2025) · International Symposium on Communications and Information Technologies · doi:10.1109/iscit67082.2025.11231686
“A significant advancement in regression-based techniques was marked by the introduction of YOLO by Redmon et al. in 2016 [11], which allowed for real-time object detection by employing a single forward pass that concurrently executes both localization and classification tasks.”
de Antioquia, U., Ramirez-Agudelo, J.-F., Bedoya-Mazo, S., et al. (2025) · Revista Colombiana de Ciencias Pecuarias · doi:10.17533/udea.rccp.e358533
Two-Stage Visual Feature Learning for RIS-Assisted Beamsteering in Occluded User Scenarios
Nishiyama, S., Chakraborty, D., & Okada, M. (2025) · International Symposium on Communications and Information Technologies · doi:10.1109/iscit67082.2025.11231550
“In contrast, single-stage detectors like YOLO [19] provide real-time performance by directly predicting object classes and bounding boxes.”
Cost of Artificial Intelligence: A Survey in Finnish Software Companies
Klemetti, A., Sorvisto, A., Raatikainen, M., & Nurminen, J. K. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-12089-2_37 · Cited by 1
DTF-Net: Dual-Transfer Feature Network for Ship Detection in Inclement Weather Conditions
Le, T.-H., & Hoang, Q.-V. (2025) · International Symposium on Communications and Information Technologies · doi:10.1109/iscit67082.2025.11231681
“YOLO family [7]–[9], one-stage detectors that formulates detection as a single regression problem, enabling real-time performance.”
Visztra, G. V., Balázs, P., Makai, Á., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5778713
Diao, H., Tang, C., Xu, B., Luo, H., Li, M., Wang, Y., & Tang, X. (2025) · doi:10.1109/iccad66269.2025.11240916
“Artificial intelligence (AI) based on neural networks (NN) has been widely applied in edge-side scenarios such as face recognition and voice interaction [1]–[4].”
Vehicle Counting for Traffic Management
K, D. A., & R, P. K. K. (2024) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-29984
Xiao, W., Hu, L., Li, Z., Wang, L., Zhang, W., Jiang, L., & Long, W. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-4045-7_48
A Hybrid Reasoning System for Visual Arithmetic Learning
Zhang, W., Johnson, C. G., & Özcan, E. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-11402-0_2
Saeaifisppf: Squeeze Aggregation Excitation Intrascale Feature Interaction Spatial Pyramid Pooling
Guo, C., & Ren, K. (2025) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss55030.2025.11243981
“One-stage object detection is a regression-based detection that locates the regression after extracting the features, and the representative algorithms are the You Only Look Once (YOLO) series [3–12] and the Single Shot MultiBox Detector (SSD) series [13, 14].”
Deep Learning for Semiconductor Chip Quality Assessment
Yang, L. I. U., Xiaozhan, L. I., ZHANG, X., Junqi, Y. U. A. N., ZHANG, B., & LIANG, M. (2025) · IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences · doi:10.1587/transfun.2025eap1108
Unidirectional guidance network for enhanced small object detection in UAV imagery
Deng, X., Xu, Z., & Yang, W. (2025) · Multimedia Systems · doi:10.1007/s00530-025-02058-z
КЛАССИФИКАЦИЯ СОСТОЯНИЯ ПТИЦЫ НАПОЛЬНОГО СОДЕРЖАНИЯ НА ОСНОВЕ ИСПОЛЬЗОВАНИЯ YOLO
Bukhanov, D. G., Chernikov, S. V., & Koshlich, Y. A. (2025) · doi:10.15622/rcai.2025.043
Modular Model Stacking Based on YOLO: Novel Approach for Model Extension Without Retraining
Chang, Y.-C., Gu, H.-Y., Wu, Z.-R., Chang, Z.-Q., Chen, L.-Z., Cheong, W. L. W., Lim, X.-Y., & Chien, W. (2025) · IEEE International Conference on Computation, Big-Data and Engineering (ICCBE) · doi:10.1109/iccbe65177.2025.11256046
“YOLOv1–v3 introduced the concept of unified detection pipelines, achieving high speed with relatively simple backbone networks [1–3].”
A mobile platform-friendly lightweight traffic light detection and recognition model
Chen, T., Han, J., Wang, J., et al. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06993-2
“Common one-stage methods are usually based on You Only Look Once (YOLO) [ 8–13 ] or Single Shot MultiBox Detector (SSD) [ 14].”
Automatic Detection and Fastening Inspection of High-Strength Bolts in Steel Bridge Construction
Inoue, M., Morita, K., & Wakabayashi, T. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-10150-1_14
Multi-view Point Cloud Object Recognition in Rehabilitation and Daily Life Intelligent Systems
Song, Q., Dou, W., Wang, W., Chyanzheng, S., & Kubota, N. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-10150-1_31
Li, Z., Yang, Y., Xie, M., et al. (2025) · Journal of Computing in Civil Engineering · doi:10.1061/jccee5.cpeng-7321
YOLO-Based Detection of Buy and Sell Signals in Candlestick Charts with Moving Averages
Santos, M. A. S., da Silva, A. R., & Ortoncelli, A. R. (2025) · doi:10.5753/eniac.2025.12471
“The first version of YOLO (YOLOv1) resizes an image to 448 × 448 pixels as input to a single CNN, directly predicting bounding boxes and classes per grid cell [Redmon et al. 2016].”
Pig Detection Using Computer Vision Models Trained on Synthetic Datasets
Martins, A. V., Paixão, T. M., Boldt, F. A., et al. (2025) · doi:10.5753/eniac.2025.14272
“The model we trained to evaluate the efficacy of the method was the YOLO model [Redmon et al. 2016], version 12 [Tian et al. 2025b].”
MAAN: Power Object Detection Network via Visual State Space Model and Area Attention
Wu, M., Wen, L., Wang, G., et al. (2025) · IEEE Access · doi:10.1109/access.2025.3640435
Argus: Quality-Aware High-Throughput Text-to-Image Inference Serving System
Agarwal, S., Mitra, S., & Iqbal, S. (2025) · International Middleware Conference · doi:10.1145/3721462.3770765 · Cited by 3
“14, we compare speed-ups across batch sizes for non-DMs such as YOLOv5n [64], ResNet50 [41], E $ cientNet-b4 [71], and GPT-8B [28], alongside various DMs .”
Optimizing Human Detection in Natural Disaster Environments Through YOLO and Image Processing
Muradkhanli, L. (2025) · IEEE International Conference on Application of Information and Communication Technologies (AICT) · doi:10.1109/aict67988.2025.11268634
“YOLO algorithm was originally proposed by Redmon et al. in 2016 as a real-time object detection framework [11].”
Blind Assistance System Using YOLOv8
Sinchana, M., U, M. S., H, M. K., Geetha, N., & Prof. Chethan Kumar, T. (2025) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-30318
Artificial Intelligence-Based Mold Defect Detection SystemPreprint
xie, C., Xie, C.-F., Chen, C.-Y., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5886013
Pu, Y., Zhang, R., & Zhang, Z. (2025) · doi:10.1145/3772128.3772168
“Additionally, single-stage detectors represented by the YOLO series [ 6 ] are renowned for their rapid iteration and significant performance improvements.”
Enhanced sodium rod detection and distribution using a YOLOv5s-SNet2-CBAM lightweight network
Song, H., Wang, J., Li, Z., et al. (2025) · The Visual Computer · doi:10.1007/s00371-025-04274-0
“The you only look once (YOLO) series [6] constitutes an outstanding model in object detection, renowned for its accuracy and rapid detection speed.”
Automated Annotation of Shearographic Measurements Enabling Weakly Supervised Defect Detection
Plassmann, J., Schuler, N., Schuth, M., & von Freymann, G. (2026) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-032-29430-2_17 · Cited by 2
Flexecutor: Out-of-the-Box Smart Provisioning for Serverless Workflows
Molina-Giménez, E., Barcelona-Pons, D., Iacoponelli, O. H., et al. (2025) · doi:10.1145/3774899.3775013
Train Wheel Tread Defects Detection Algorithm Based on SEN-YOLO
You, L., Mo, Y., Tu, J., & Zhou, H. (2025) · SAE technical papers on CD-ROM/SAE technical paper series · doi:10.4271/2025-99-0458
Guo, Y., Jin, J., Sun, G., & Liu, Y. (2025) · doi:10.1145/3761668.3761683
“In contrast, the development of deep learning-based methods [14], especially the YOLO series of algorithms [15, 16], provides new opportunities for improving the accuracy of curling stone detection.”
Robotic Removal of EV-LIB-Pack LIDS
Islam, A., Nenadic, N., & DeHaven, M. (2025) · doi:10.65569/uxib2166
Multimodal Collaborative Decision-Making Risk Detection Technology for Substations
Guo, L., Tian, Y., Wang, Y., et al. (2025) · doi:10.1145/3761668.3761704
“YOLOv8 algorithm has become the current benchmark framework for target detection due to its outstanding advantages in detection accuracy and speed [14].”
ClearFace: Facial Acne Detection and Classification System Using YOLOv11 and EfficientNet-B0
Viana, T. A. D. S., Gonçalves, I. H., de Paula, I. S., Silva, F. H. D. S., & Filho, P. P. R. (2025) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.5753/sibgrapi.est.2025.38285 · Cited by 1
“…general but less accurate in medical contexts—and SSD (Single Shot Multibox Detector) which tends to be more precise for medical applications but also slower are also frequently applied in the field, even though they were not originally designed specifically for facial lesion detection [44], [45].”
Detection of Retinal Anatomical Structures and its Application to Image Quality Assessment
Michelassi, R. D. C., & Hirata, N. S. T. (2025) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.5753/sibgrapi.est.2025.38309
“We chose the state-of-the-art YOLO model [7] for the detection tasks.”
An Improved YOLOv8 Algorithm Based on BiFPN Architecture for Detection of Aerospace Connector Pins
Wang, C., Liu, S., Ni, Z., et al. (2025) · International Symposium on Aerospace Engineering and Systems (ISAES) · doi:10.1109/isaes66870.2025.11274381
“In view of this, many single-stage detection algorithms have been proposed, such as SSD [7] and YOLO [8] .”
Nunes, G., & Rodrigues, P. S. (2025) · doi:10.5753/sibgrapi.est.2025.38275
“Unlike YOLO [31], Faster R-CNN handles small object detection more effectively, achieving processing rates of up to 5 FPS.”
Visual Identification System for Aedes Detection using Drones
Soares, H., do Patrocínio, Z. K. G., Jr., & Guimarães, S. J. F. (2025) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.5753/sibgrapi.est.2025.38303
Sistema de Detecção e Classificação de Resíduos Sólidos com Computação em Borda Usando Deep Learning
de Andrade, A. L. L., de Almeida, C. H. O., Rodrigues, D. D. A., et al. (2025) · doi:10.5753/sibgrapi.est.2025.38287
Estimation Time of Escape (ETE) by Monitoring the Building Evacuation Flow
Ostermann, M., Akin, E., & Sarkheyli-Hägele, A. (2025) · International Conference on the Internet of Things (IoT) · doi:10.1145/3770501.3770517
“You Only Look Once (YOLO) [17] is an object detection model recently proposed in 2016/2017 to improve the object detection field by balancing speed and accuracy.”
Semantic-integrated multi-model fitting for real-time VSLAM in highly dynamic environments
Zhang, T., Li, N., Gong, G., & Lin, X. (2025) · The Visual Computer · doi:10.1007/s00371-025-04235-7 · Cited by 1
“This technique is pluggable, as it is compatible with various off-the-shelf detectors (whether one-stage [26], two-stage [27], or anchor-free [28]), and is zero cost in terms of additional training.”
METHODICAL INVESTIGATION OF LOSS FUNCTIONS IN DEEP LEARNING AND MACHINE LEARNING
Anbalagan, T., & Nath, M. K. (2025) · Biomedical Engineering Applications Basis and Communications · doi:10.4015/s101623722530007x
Open-Vocabulary Multi-Object Tracking Based on Multi-Cue Fusion
Xu, L., Bai, J., Nai, L., & Liu, C. (2025) · Applied Sciences · doi:10.3390/app152413151
Marques, N., Neves, G., Carreira, C., et al. (2025) · Proceedings - International Conference on Tools with Artificial Intelligence, TAI · doi:10.1109/ictai66417.2025.00096
“YOLO-based object detectors [2] are widely used in industrial defect detection due to their speed and accuracy.”
LSL-ISTD: A local saliency-aware learning framework for infrared small target detection
Xi, Y., Liu, D., Liu, Z., Kou, R., Wu, Y., & Zhang, J. (2025) · Infrared Physics & Technology · doi:10.1016/j.infrared.2025.106322 · Cited by 5
VIP: A visual insulated pipeline dataset for computer vision tasks
Papadam, D. R., Zamioudis, A., Mentesidis, P., Charalampakis, E., Valsamara, I., Spatharis, E., Mygdalis, V., & Pitas, I. (2025) · Journal of Pipeline Science and Engineering · doi:10.1016/j.jpse.2025.100421
Forest Fire Monitoring from Unmanned Aerial Vehicles Using Deep Learning
Graveline, C., & Payeur, P. (2025) · International Electronic Conference on Sensors and Applications (ECSA) · doi:10.3390/ecsa-12-26597 · Cited by 1
YOLO-SAD: Enhancing Small Arthropod Target Detection for Autonomous Port Inspection Robots
Song, B., Zi, P., Pei, J., et al. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3643408
“…years, compared with two-stage approaches such as the R-CNN family of approaches [7], [8], [9], single-stage detection algorithms, such as SSD [10] and the YOLO series [11], [12], [13], [14], [15], [16], [17], [18], [19], have demonstrated superior speed and efficiency in dynamic object detection.”
YSBE-Based Target Detection via Multibeam Sonar on Velocity-Constrained AUVs: Toward Port Inspection
Yang, X., Yu, S., Zhu, J., et al. (2025) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2025.3609203
“Recent years have witnessed substantial improvements in deep learning-based detection frameworks [10], [11], [12], [13], [14], with the you only look once (YOLO) architecture emerging as a dominant paradigm due to its exceptional balance of detection speed and precision.”
Drone-Based Coastline Pollution Monitoring System for Detecting and Collecting Garbage
Tan, C. J., Janthori, T., Hayashi, E., et al. (2025) · Journal of Robotics and Mechatronics · doi:10.20965/jrm.2025.p1374
Computer Vision for Autonomous Driving
Suguna, A. (2025) · International Journal of Innovative Research in Information Security · doi:10.26562/ijiris.2025.v1112.03
Highway Animal Migration Monitoring and Statistical System Based on Infrared Imaging and YOLOv8
靳, 辰. (2025) · Software Engineering and Applications · doi:10.12677/sea.2025.146112
Improving GUI Element Detection with a Refined YOLO11-Based Approach
Gao, Z. (2025) · International Conference on Electronic Information Technology and Computer Engineering (EITCE) · doi:10.1145/3766671.3766693
“Its core idea involves modelling object detection as a regression problem, simultaneously predicting bounding box locations and class probabilities through a single forward propagation, thereby optimising the end-to-end detection workflow [2].”
IBBI: A Python package for the detection and classification of bark and ambrosia beetles
Marais, C., Kuo, E., Hulcr, J., et al. (2025) · The Journal of Open Source Software · doi:10.21105/joss.08796
“This is ideal for processing raw images from traps or collection sheets and is based on fine-tuned high-performance architectures like YOLO (Redmon et al., 2016) and detection transformers (Lv et al., 2023).”
Real-Time Child Detection Using YOLO Architectures for Enhanced Safety Applications
Alsamaani, A., Almakki, R. S., & Obidallah, W. J. (2025) · IEEE International Conference on e-Business Engineering (ICEBE) · doi:10.1109/icebe68123.2025.00047
“YOLO was developed by Joseph Redmon and Ali Farhadi in 2015 which stands for ”You Only Look Once” [6].”
Ahmed, S. T., Akshaya, K. R., Vattikuti, H., Preetham, L. S. P., & Dutta, R. (2025) · International Conference on Vehicular Technology and Transportation Systems (ICVTTS) · doi:10.1109/icvtts67119.2025.11296481
“The field of traffic monitoring has evolved significantly, transitioning from hardware-reliant systems to advanced AI-driven frameworks powered by computer vision and deep learning [1].”
Computer Vision-Based Pose Estimation for Student Engagement Detection: Trends and Challenges
Tarigan, G. A., Elwirehardja, G. N., Nugroho, K. S., & Pardamean, B. B. (2025) · International Conference on Cybernetics and Intelligent System (ICORIS) · doi:10.1109/icoris67789.2025.11296082
“3, the most commonly used method for pose detection is object detection, utilizing variations of YOLO [12], including YOLOv3 [13], [14], [15], YOLOv4 [16], and YOLOv5 [17], [18], which are featured in five studies.”
Identification of tumoral region in CT images of cervical cancer using neural networks
Waldo-Benítez, G., Parra-Sánchez, L. F., Padierna, L. C., et al. (2025) · AIP conference proceedings · doi:10.1063/5.0310194
Development of a Program for Process Monitoring in Production Systems
Pechenin, V. A., Kovaleva, A. M., Kiseleva, P. I., et al. (2025) · Journal of Machinery Manufacture and Reliability · doi:10.1134/s1052618825700839
“The second method for detecting parts is to use methods based on the YOLO family of neural networks for detection [10].”
Improved YOLOv8-based crack detection model for road bridges
Li, X., Chen, L., Zhang, H., et al. (2025) · doi:10.1109/isemc68048.2025.11291860
“…which has been widely used in the field of crack recognition by abandoning the region extraction step and solving the detection task as a regression problem, directly obtaining the location and category information of the cracks in the image, and improving the efficiency of crack recognition [5].”
N, A. C., & Thiruppathi, S. (2025) · International Conference on Vehicular Technology and Transportation Systems (ICVTTS) · doi:10.1109/icvtts67119.2025.11296504
“YOLO models, as described in [5] and [6] are more efficient and designed for resource-constrained devices, which accounts for the better speed.”
Girardet, J., Glotin, H., Chavin, S., Poupard, M., Guiderdoni, J., & Sarano, V. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103564
ParkTrack AI: Hybrid Vision System for Real-Time Parking Space Detection
Kurakula, V., Kharthik, B., & Minchula, V. K. (2025) · International Conference on Cyber Physical Systems, Power Electronics and Electric Vehicles (ICPEEV) · doi:10.1109/icpeev67897.2025.11291447
“Simultaneously, object detection frameworks such as Faster R-CNN [4], SSD [5], and YOLO [6] facilitated real-time vehicle detection.”
Q-Tracking: A Robust Visual Human Following for Quadruped Robots in Dynamic Environments
Su, Y., Cun, C., Xia, H., et al. (2025) · International Conference on Advanced Robotics and Mechatronics (ICARM) · doi:10.1109/icarm65671.2025.11293732
“We develop a YOLO [10]-Tracking-ReID cascaded architecture that integrates high-frequency detection with lightweight feature matching, achieving robust identification even under partial occlusions or extreme viewing angles.”
DBSD: a Fast Visual Detection Algorithm for Maritime Distressed Personnel
Yan, D., & Li, X. (2025) · International Conference on Advanced Robotics and Mechatronics (ICARM) · doi:10.1109/icarm65671.2025.11293535
“A representative example of the two-stage methods is Faster R-CNN [10], whereas among one-stage detectors, the YOLO series [11], [12] has become particularly prominent for its balance between accuracy and inference speed.”
DRACO: Depth-Assisted Robot ACtOr with Language-Conditioned 3D Flow Matching Policy
Ukarapol, T., Li, X., & Liu, H. H. (2025) · International Conference on Advanced Robotics and Mechatronics (ICARM) · doi:10.1109/icarm65671.2025.11293556
Subjective Assessment of Operator Responses for Mobile Defect Identification in Remanufacturing
Ahmed, A., Mohandas, R., Joorabchi, A., et al. (2025) · Irish Signals and Systems Conference · doi:10.1109/issc67739.2025.11290000
“Machine vision systems employing camera-acquired imagery [1] have automated defect detection through deep neural networks such as YOLO [2] and Faster R-CNN [3] achieving superior performance over conventional classifiers like support vector machines.”
Development of Distance Measurement Estimation Model for Specific Objects
Herdianto, H., Sihombing, P., Fahmi, F., & Tulus, T. (2025) · International Conference on Cyber and IT Service Management (CITSM) · doi:10.1109/citsm67730.2025.11291500
“Therefore, several studies have been conducted on the ability of soccer robots to detect objects, including [3], [4], [5], [6], [7], [8], [9] and especially measuring the distance of objects using cameras, have been done by previous researchers, such as those conducted by [1], [2], [10], [11],…”
TerraVerse: A Multimodal Sensor Dataset for Unstructured Scenes Spanning Multiple Regions in China
Zhang, X., Tan, Y., Wang, B., & Liu, H. (2025) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus66297.2025.11295498
“We used Open Labeling [17] to annotate the image targets, and the annotation results were in YOLO format [18], with a total of more than 160K RGB images annotated, as shown in Fig.”
RRAM-Based Spectral-Domain Convolution Accelerator for Reliable and Energy-Efficient CNN Inference
Wen, J., Baroni, A., Wenger, C., Krstic, M. D., & Poehls, L. M. B. (2025) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2025.3644141
FAMD: A Fusion-Aided Multimodal Object Detection Framework
Du, Y., & Zhu, J. (2025) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus66297.2025.11294942
“While modern detectors, such as YOLO-series models [1], [2], [3], [4] and the two-stage detector family [5], [6], have achieved remarkable accuracy under controlled settings, their performance often deteriorates in real-world environments due to their reliance on single-modal visual data.”
Liu, Y., Wang, H., Wen, T., et al. (2025) · doi:10.1109/powercon66300.2025.11294047
Zhang, Y., Hei, Y., Duan, X., Zhang, W., Lin, C., & Zhou, J. (2025) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus66297.2025.11295715
“A significant breakthrough occurred in 2016 when You Only Look Once (YOLO), a single-stage object detection model developed by J.Redmon, was proposed [5].”
Biouki, M. A., Soheili, F., Delfan, N., Masoudifar, N., Mohaghegh, N., Ebrahimi, S., Daneshvar, S. M. M. H., Tahernezhad, Y., Sun, C., Glogauer, M., et al. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3647413
“In contrast, one-stage detectors—such as YOLO and SSD— perform detection in a single step, eliminating the need for region proposals and achieving much faster inference [55], [56].”
Find My Meow: A Platform for Locating Lost Cats
Sookying, J., Wiangin, K., & Ratanaworabhan, P. (2025) · International Computer Science and Engineering Conference · doi:10.1109/icsec67360.2025.11298094
Ai Virtual Inspection and Environmental Monitoring Platform
Jiawen, G., Azman, A., Jazri, H., Sangodiah, A., & Hussain, K. (2025) · International Conference on Emerging Trends in Networks and Computer Communications (ETNCC) · doi:10.1109/etncc66224.2025.11299743
Baek, J.-W., Jeong, U.-C., Kim, B.-M., Kim, Y.-J., & Kim, J. (2025) · Journal of the korean society of manufacturing technology engineers · doi:10.7735/ksmte.2025.34.6.414 · Cited by 1
3D Reconstruction and Instance Segmentation of Shiitake Mushrooms for Accurate Volume Estimation
Ishizuka, T., Matsui, M., Sasaki, Y., et al. (2025) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc266238.2025.11293346
“Among the many approaches explored over the past decade, a landmark contribution is the “You Only Look Once” (YOLO) [9]–[12] framework, first introduced by Redmon and colleagues.”
Study of Small Target Defect Detection on Packaging Surface Based on Improved YOLOv5n
Jiang, S., Zhang, Z. P., Wang, F., Li, Z., Lu, Y., & Liang, J. (2025) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus66297.2025.11295449
“YOLOv4 proposed BoS strategy and BoF strategy to optimize and improve the training phase [22].”
Research on Traffic Sign Detection Algorithm Based on Improved YOLO11n
Feng, H., Meng, J., Guo, Z., Zhao, P., Zhang, W., Cao, Y., & Liang, C. (2025) · Technologies · doi:10.3390/technologies14010004 · Cited by 2
Optimized YOLOv8–DenseNet Pipeline for Fast Cervical Cell Detection
Tachaanantakit, T., & Rattanatamrong, P. (2025) · International Computer Science and Engineering Conference · doi:10.1109/icsec67360.2025.11297990
ROS Based Multi-Person 3D Pose Extraction for Edge Deployment
Bampos, K., Piperigkos, N., Anagnostopoulos, C., et al. (2025) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc266238.2025.11293258
“In selecting the appropriate 2D keypoint detector for real-time deployment, we evaluated the computational complexity of Mask R-CNN [13] and mid-range YOLO variants [14].”
Γκίλλας, Α., Anagnostopoulos, C., Piperigkos, N., et al. (2025) · IEEE International Smart Cities Conference (ISC2) · doi:10.1109/isc266238.2025.11293277
“A YOLOv8-s model [17], [18], comprising approximately 11M parameters, is used to detect both passengers and common items (e.g., phones, backpacks, laptops).”
Intelligent Mobile Robot Fall Detection System Integrating Pose Estimation and YOLO
Luo, Y.-Y., Cheng, C.-H., & Chiu, T.-Y. (2025) · International Conference on Awareness Science and Technology (iCAST) · doi:10.1109/icast68191.2025.11300024
“YOLO (You Only Look Once) [4] As illustrated in Fig.”
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.