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 47,701–47,800 of 51,398
A Systematic Method for Identifying Forged Brand Logos
Sathyamurthy, E., Shrivastav, P., Sabarinath, S., et al. (2025) · doi:10.1109/iccist67338.2025.11438776
Real-Time Railway Maintenance Inspection on Embedded Devices
Erdeniz, F. S., & Küçükmanіşa, A. (2025) · International Conference on Embedded Systems and Artificial Intelligence (ESAI) · doi:10.1109/esai67033.2025.11438536
“These models predict the locations and classes of objects simultaneously [3].”
Enhanced infrared ship detection via AAF-YOLO: a lightweight and accurate approach
Li, Z., Zhang, D., Xu, M., Su, Y., & Xu, Z. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08419-3
Intelligent safety monitoring algorithm for mining conveyor belts based on lightweight deep learning
Ren, Q., Wu, L., & Zhang, M. (2026) · Advances in Mechanical Engineering · doi:10.1177/16878132261432109
“8 12 In prior studies, 13 scholars extensively explored the significance of utilizing mining conveyor belts.”
Deng, G., Li, Z., & Gao, Z. (2026) · Scientific Reports · doi:10.1038/s41598-026-45083-2
“…approach of detecting wheat, which involves manually constructing features based on the analysis of wheat’s shape and color, the neural network autonomously constructs features of wheat by assimilating a vast amount of wheat data and then applies them to the detection of wheat 4,6,7,19,21–24 .”
Saipattalung, K., Vorapracha, P., Thongyoo, T., et al. (2026) · International Journal of Electrical and Electronics Research · doi:10.37391/ijeer.140113
Website Segmentation Beyond Structure: A Benchmark on Functional and Digital Maturity Classes
Saxer, J., Gerber, J., Weiler, A., & Grossniklaus, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-21300-6_14
Li, Y., & Feng, L. (2026) · International Conference on Embedded Systems, Mobile Communication and Computing (EMC²) · doi:10.1109/emc68537.2026.11441937
“In contrast, one-stage methods — such as the YOLO series [9] and SSD [10] — execute object localization and class prediction concurrently within a single forward pass, thereby demonstrating significant advantages in inference efficiency.”
Virtual Testbed for Ai-Driven Driver Monitoring Systems in Electric Vehicles
Ambulkar, A., & Deshpande, S. J. (2025) · IEEE International Transportation Electrification Conference (ITEC-India) · doi:10.1109/itec-india67626.2025.11439614
“Convolutional neural networks (CNNs) have demonstrated superior performance in facial landmark detection and drowsiness classification [6] framework for ADAS validation, though their approach lacks specific DMS modules.”
Lightweight Object Detection Network Based on Multi-Scale Feature Fusion
Xu, C. (2025) · IEEE International Conference of Safe Production and Informatization (IICSPI) · doi:10.1109/iicspi66775.2025.11438069
“Object detection, a fundamental technique in computer vision, aims to accurately identify the categories of target objects in images or videos while pinpointing their spatial positions [1, 2] .”
Small Target Detection in Remote Sensing Images Based on Multiscale Contextual Enhancement
Yang, J., Chen, G., & Yao, Y. (2025) · IEEE International Conference of Safe Production and Informatization (IICSPI) · doi:10.1109/iicspi66775.2025.11438215
“In 2014, deep learning was first introduced into the field of target detection, which realized a revolutionary improvement in detection accuracy[1].”
Yang, R., & Zhang, Q. (2025) · doi:10.1109/aibdf67964.2025.11440758
“With the advancement of the YOLO [10] series, these algorithms have been widely applied across various fields.”
Cao, Y., Nie, Z., Sun, F., et al. (2026) · Composites Communications · doi:10.1016/j.coco.2026.102800
Optimizing Tomato Object Detection in Smart Agriculture with Attention-Augmented AI Vision Systems
Sahlaoui, H., Alaoui, E. A. A., & Agoujil, S. (2025) · doi:10.1109/esai67033.2025.11438635
“…catalyzed the shift toward single-stage detectors, with the YOLO family emerging as the predominant choice due to its superior speed-accuracy balance, as demonstrated in large-scale studies by Redmon et al. [14] showing consistent 30+ FPS performance while maintaining competitive accuracy.”
Real Time Facial Emotion Detection with an Expo Go Mobile Interface
Sribharathi, B., A, J. K., S, M. M., et al. (2026) · doi:10.1109/icauc68182.2026.11441215
“Most face recognition systems are still not very good at working on devices or they need a lot of special equipment [15].”
Indian Traffic Sign Dataset for Self Driving Cars in YOLO Format
Sane, P., Joshi, V. S., & Patel, S. (2025) · IEEE International Transportation Electrification Conference (ITEC-India) · doi:10.1109/itec-india67626.2025.11439261
“Initially introduced by Redmon et al., YOLO’s single-shot detection approach enables both classification and localization within a cohesive framework, resulting in notable enhancements in the detection of various objects, including traffic signs [8], [12].”
Leveraging Neurosymbolic AI for Slice Discovery
Collevati, M., Eiter, T., & Higuera, N. (2026) · Neurosymbolic Artificial Intelligence · doi:10.1177/29498732261419315
Vision-LLM for Intelligent 6G Vehicular Networks
Akilandeswari, J., Kumar, A. S., Kumar, A. N., et al. (2026) · doi:10.1109/iccces62661.2026.11436734
“Initial studies on real-time object recognition, like the project by Redmon et al. [9], built the basis of fast and unified algorithms in detecting objects, whereas measures of structural similarity, suggested by Wang and colleagues [3], can be utilized to evaluate the quality of images, and…”
Tan, Z., Tan, Y., Lin, P., Su, W., He, T., & Wu, W. (2026) · Sensors · doi:10.3390/s26072027
Semantic-Aware Adaptive Denoising for Robust Nighttime Vehicle Detection and Tracking
Sun, T. R., Liu, X., Fu, H., et al. (2025) · International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc68654.2025.11438079
“Recent advances in deep learning have improved detection robustness through architectures like Faster R-CNN [5] and YOLO [6], yet their performance at night remains limited by uniform denoising strategies that often blur critical features.”
Depth of Reinforcement Learning Optimization of Low Traffic Conflict Resolution Model
Liu, L., Shi, X. S. X., & Zhang, F. (2025) · IEEE International Conference on Electrical, Automation and Computer Engineering (ICEACE) · doi:10.1109/iceace67491.2025.11439740
YOLO-Based and DCNN Fusion for Urban Traffic Congestion Prediction
Chen, Y. (2025) · IEEE International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc68654.2025.11438233
“The model is also with high real-time and excellent performance of detecting bloster and small objects, which can capture reliably urban traffic scenarios’ dense sma ll vehicle targets, and meet research requirements [8].”
GBSA-YOLOv8: A Multi-Scale Real-Time Detection Model for Rice Field Pests in Complex Field Scenarios
付, 欣. (2026) · Artificial Intelligence and Robotics Research · doi:10.12677/airr.2026.152059
An Improved YOLOv11-Based Method for Weld Defect Detection
Liu, W., Zhang, F., & Wang, X. (2025) · IEEE International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc68654.2025.11437929
“Among various deep learning models, the YOLO (You Only Look Once) series [7]–[10] has demonstrated great potential in industrial defect detection due to its end-to-end architecture, high inference efficiency, and excellent real-time performance, and has gradually been applied to weld defect…”
Detection of Dental Restorations in Digital Panoramic Radiographs Using YOLOv11
Aydın, F., Ayran, Ş., & Yuce, M. K. (2026) · European Annals of Dental Sciences · doi:10.52037/eads.2026.0007 · Cited by 1
Zhu, L., & Long, W. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08433-5 · Cited by 1
Han, B. H., Xu, R., Peng, L., et al. (2025) · International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc68654.2025.11437646
“…to efficiently accomplish complex tasks (e.g., object recognition, semantic segmentation) with robust local explicit semantic capture capability [9]; while Swin Transformer (SwinT) [10], a leading Transformer variant in computer vision, employs its unique Shifted Window based Multi-head…”
Focus Track: A Multi-Faceted AI-Powered Framework for Real-Time Attention and Distraction Monitoring
Kiruthik Kumar, J., K, S. S. S., Thiruvarshan, M., A, M. N. A., Sri Pavithra, B., & George, S. V. (2026) · International Conference on Communication, Computing and Electronics Systems (ICCCES) · doi:10.1109/iccces62661.2026.11436287 · Cited by 1
“This was followed by deeper architectures such as VGG, which significantly enhanced image classification and representation learning through increased network depth [7].”
Zeng, B., Zheng, X., Zhang, L., Jin, X., Huang, S., Zhang, Y., Hu, S., & Tian-Ge, W. (2025) · IEEE International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc68654.2025.11437972
“The YOLO [17] family of detectors has gained popularity due to its balance between accuracy and computational efficiency, making it suitable for real-time monitoring applications.”
Design of Tobacco Leaf Information Detection Device Based on Target Detection Model
Lei, J., Chen, Z., Cheng, S., et al. (2025) · doi:10.1109/actce66599.2025.00138
“Leading-edge first-stage object detection algorithms such as YOLO [9] and SSD [10] strike an optimal balance between accuracy and computational efficiency, making them particularly suitable for real-time industrial inspection scenarios.”
Qin, J., Ding, H., & Wu, T. (2026) · Geomorphology · doi:10.1016/j.geomorph.2026.110299
Dong, R., Wei, X., Zheng, D., et al. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6470331
Ferdaus, M. M., Dam, T., Sarkar, M. R., & Anavatti, S. G. (2026) · Advanced Engineering Informatics · doi:10.1016/j.aei.2026.104598 · Cited by 1
Bhattarai, S. (2026) · Research Square · doi:10.21203/rs.3.rs-9212377/v1
A Lightweight RGB-D Object Detection Model with Multi-teacher Adaptive Knowledge Distillation
Wu, X., Zhang, B., & Wan, W. (2026) · International Journal of Automotive Technology · doi:10.1007/s12239-026-00448-4
Uncertainty-Aware License Plate Recognition for Data-Driven Urban Traffic AnalyticsPreprint
Bhattarai, S. (2026) · Research Square · doi:10.21203/rs.3.rs-8900771/v1
Research on Sensory Recognition System Design for Exhibition Design Based on Artificial Intelligence
Hou, Z., Zhou, C., & Wang, Z. (2025) · IEEE International Conference on Electrical, Automation and Computer Engineering (ICEACE) · doi:10.1109/iceace67491.2025.11439513 · Cited by 1
Research on Vehicle Detection and Tracking Algorithm Based on Improved YOLOv5s
Xie, L., Dong, X., & Li, X. (2025) · doi:10.1109/actce66599.2025.00133
“Redmon et al. [1] first proposed solving the algorithm detection task as a regression problem, where a single CNN network processes the image and directly predicts both the object category and location coordinates.”
Luo, W. (2025) · IEEE International Conference on Electrical, Automation and Computer Engineering (ICEACE) · doi:10.1109/iceace67491.2025.11439558
“Deep learning-based approaches, particularly YOLO series models [7-10], have become dominant for their superior accuracy and speed.”
Vehicle Classification and Accident Detection Based on the Fine-tuned YOLOv10
Huang, Z., & Mao, M. (2025) · doi:10.1109/actce66599.2025.00089
“Our experiments show that YOLO [4] models are generally well-suited for vehicle classification and accident detection, with YOLOv10 [5] demonstrating the best overall performance: • Highest Accuracy : The YOLOv10x achieves the highest accuracy across all metrics.”
InfluentialMSA-DETR: Multi-scale attention enhanced DETR for object detection in oilfield surveillance
Cao, Q., Ma, J.-R., & Zhang, L.-B. (2026) · Petroleum Science · doi:10.1016/j.petsci.2026.03.045
Bulut, A., Aşkın, M., & Cinarer, G. (2026) · Diagnostics · doi:10.3390/diagnostics16070977 · Cited by 1
“In addition, object detection and segmentation frameworks, such as Faster R-CNN [19], YOLO [20] and U-Net based architectures [21], extend CNN capabilities beyond classification by enabling simultaneous localization and structural analysis.”
Copper Foil Surface Defect Oriented Long-Tailed Object Detection
Xia, Z., Zhang, T., Yang, D., et al. (2025) · International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc68654.2025.11438129
“Recent advances in convolutional neural networks (CNNs) have driven significant progress in object detection, with emerging algorithms such as the YOLO series [1-4], RT-DETR [5], and EfficientDet [6] achieving remarkable balance between detection accuracy and inference speed.”
3-D SPGNet: A 6-DoF Grasp Detection Network via 3-D Surface Constraint and TSDF Reconstruction
Yu, H. B., Zhang, X., Zhao, Z., & Chen, H. (2026) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2026.3672807
“In fact, similar to bounding boxes generated in object detection [17], [18], [19], grasp detection models predict grasp candidates for a given scene, where each candidate is associated with a grasp score.”
Zhang, W., Guo, C., Guo, J. Q., Du, B., & Guo, S. (2026) · Engineering Research Express · doi:10.1088/2631-8695/ae50cb
Context Dropout: A Novel Method to Reduce Contextual Reliance in Object Detection
Vonderhaar, L., Elvira, T., Procko, T., Newcomb, A., & Ochoa, O. (2025) · IEEE International Conference on Future Machine Learning and Data Science (FMLDS) · doi:10.1109/fmlds67896.2025.00042
“In more recent designs of Object Detection Models (ODM), such as You Only Look Once (YOLO), background context may be used to predict the presence of objects much quicker and more efficiently than without [1].”
Intense Punch Exchange Detection in Boxing Videos Using Pose-Based Machine Learning Pipeline
Pitigala, P. K. D. G. D., & Ranathunga, L. (2026) · IEEE International Conference on Industrial and Information Systems (ICIIS) · doi:10.1109/iciis69028.2026.11450926
“The YOLOv8n (nano) architecture was selected as the base model due to its optimal balance between detection accuracy and computational efficiency [12].”
SODNet: a scale-oriented detection network for efficient UAV-based sewage outfall detection
Zeng, L., Liu, X., Dai, B., Wen, L., & Du, Z. (2026) · Scientific Reports · doi:10.1038/s41598-026-45595-x · Cited by 1
Li, J. (2026) · Scientific Reports · doi:10.1038/s41598-026-44473-w · Cited by 3
“Modern detectors based on region proposal networks handle varying camera angles and lighting conditions 19 , though they occasionally struggle when hands overlap.”
Raghuwanshi, P., Rifat, A., Sathwik, G., et al. (2025) · doi:10.1109/etcom66606.2025.11436744
“One of the best objects to be used in real-time applications is YOLO, as introduced by Redmon et al. in [5].”
Detección de Enfermedades Fúngicas en taxo mediante segmentación con YOLOv11
Muñoz-Espinoza, M. S., Castillo, W. E. M., Zúñiga-Moreno, J. A., et al. (2025) · Revista Científica Multidisciplinar G-nerando · doi:10.60100/rcmg.v6i1.794
Ensemble Learning Using YOLO Models for Semiconductor E-Waste Recycling
Zhou, X., & Agaian, S. С. (2026) · Information · doi:10.3390/info17040322 · Cited by 3
“It is worth noting that the original source code repositories for YOLOv1 and YOLOv2, as cited in their respective papers by Redmon et al. [8,9], are no longer publicly accessible at the links provided.”
Taniguchi, K., & Yamane, T. (2026) · Japanese Journal of JSCE · doi:10.2208/jscejj.25-22008
Performance Comparison of YOLO Architectures for Lung Cancer Detection in Medical Imaging
Bhatt, S. D., & Astik, M. B. (2025) · doi:10.1109/sisimpact67725.2025.11439832
“The object detectors, namely You Look Only Once (YOLO) [14], YOLOv2 [15], YOLOv3, and YOLOv4 are the most prominent models for one-stage object detection.”
Yadav, R., Mitra, A., & Kala, R. (2026) · IEEE International Conference on Industrial and Information Systems (ICIIS) · doi:10.1109/iciis69028.2026.11450819
Shape-Attentive YOLOv12: A Novel Framework for Enhanced Obstacle Detection
Li, B., & Wang, C. (2025) · doi:10.1109/bddm68348.2025.11442146
“They have done this through several cycles of reworking the technology to establish a slow balance between accuracy and speed [3] .”
InfluentialStreamAlloc: A Framework for Analyzing and Transforming CUDA Code to Enable Asynchronous Execution
Basu, S. K., & Vedurada, J. (2026) · ACM Transactions on Programming Languages and Systems · doi:10.1145/3799893
“…play a crucial role in accelerating computation-intensive tasks across diverse fields [Johnson et al., 2019; Mittal and Vaishay, 2019; OpenAI, 2023; Redmon et al., 2016; Zhou et al., 2018], including Large Language Models, recommendation systems, computer vision, scientific simulations,…”
A Spatial-Frequence Semantic-Detail Network for Remote Sensing Object Detection
Fan, Y., Zhang, M., Yu, D., et al. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3678053
“Key algorithms include the YOLO [17] series, first proposed by Redmon et al. [16], and the single shot MultiBox detector (SSD) algorithm introduced by Liu et al. [18].”
Multimodal Smart Video Surveillance System for Real-Time Detection of Anomalous Activities
Badhusha, M., Singh, Y., Kumar, A., Arun, & Darshan, B. (2025) · IEEE International Conference on Emerging Trends in Computing and Communication (ETCOM) · doi:10.1109/etcom66606.2025.11436593
He, Z., & Chen, X. (2026) · World Electric Vehicle Journal · doi:10.3390/wevj17040177
Augmented Reality Visual Retrieval for Object Detection and Corpus-Guided Content Generation
Taheritajar, A., Zhao, J. L., & Orlosky, J. (2026) · IEEE International Conference on Artificial Intelligence and eXtended and Virtual Reality (AIxVR) · doi:10.1109/aixvr67263.2026.00015
A Lightweight Fire and Smoke Detection Model via Multi-Source Pre-Training for Real-World Scenarios
Yu, Z., Wang, C., Li, G., et al. (2026) · doi:10.1109/iceaai68945.2026.11442355
“In recent years, convolutional neural network–based object detectors, especially YOLO (You Only Look Once) series models, have demonstrated promising performance in fire and smoke detection tasks owing to their balance between accuracy and efficiency [4], [5].”
Road Safety for Everyone: An AI-enabled Pothole Detection System
Bamitale, A. A., Kumar, P., Maple, C., et al. (2026) · International Conference on Consumer Electronics · doi:10.1109/icce67443.2026.11449831
“…image acquisition as a foundational sensing mechanism within Emerging Consumer Technology (ECT), transforming smartphones into distributed infrastructure-monitoring nodes integrated with real-time deep-learning–based pothole validation using the You Only Look Once (YOLO) architecture [10]–[12].”
Fault Detection in Transmission Line Insulator Images with ID-YOLOv11n
Xu, W., Shao, Z., & Zhou, Y. (2025) · International Conference on Clean Energy Storage and Power Engineering (CESPE) · doi:10.1109/cespe68033.2025.00095
“Insulators, serving dual functions of electrical insulation and mechanical support in distribution lines, play a decisive role in preventing line failures and large-scale blackouts [1].”
Real-time tiny object detection in UAV aerial images with multi-scale attention fusion
Gao, J., Zhang, Y., Fan, Y., & Tian, B. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01873-5 · Cited by 1
Угрин, Д. І., Доренський, О. П., Ushenko, Y., et al. (2026) · Central Ukrainian Scientific Bulletin Technical Sciences · doi:10.32515/2664-262x.2026.13(44).33-40
Research on Inverse Kinematics Solution of UR5 Robotic Arm Based on Improved Genetic
Zhang, Y., Yu, Y., He, Y., et al. (2026) · doi:10.1109/iceaai68945.2026.11442620
“With its high flexibility, precise operation ability and strong environmental adaptability, the robotic arm has become an indispensable key equipment in multiple fields such as automated production lines, medical surgeries [1], and logistics warehousing.”
Hybrid Deep Learning Model Combining YOLOv11 and RT-DETR for Precision Weed Detection
Shivaram, B., Ramesh, T. K., & Kumar, J. (2026) · International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS) · doi:10.1109/iciscois62701.2026.11447549
“Maize, soybean Balanced accuracy and inference speed; suitable for robotics [27], [28] YOLOv3 / YOLOv4 frameworks Rice, maize Enabled real-time detection; effective for robotic actuation [19], [31] YOLOv8, YOLOv9, YOLOv10, and Faster R-CNN”
Evaluating and Enhancing YOLOv8’s Soft Error Resilience
Yang, A., Zhi, L., Guan, Y., Tan, T., & Shen, J. (2026) · Electronics · doi:10.3390/electronics15071404
Fruit quality detection method via improved YOLOv8
Li, J., & Li, L. Q. (2026) · doi:10.1109/icpege67691.2026.11451576
“In this study, the performance of the detection model is primarily evaluated utilising three metrics: precision (P), recall (R), and mean average precision (mAP)[10].”
Function-as-a-Service Empowered Edge-Cloud Analytics System for Short Video Platforms
Li, Y., Lin, K., Yuan, J., & Hu, C. (2026) · IEEE Transactions on Computers · doi:10.1109/tc.2026.3678627
“To develop suitable abstractions, we focused on five representative video analytics applications that can be applied to short video platforms: YOLO [13], Faster-RCNN [14], HOG [15], Lightroom [16], and Premiere Pro [17].”
Ciou, S.-E., Zhao, X.-L., Guo, J. C., et al. (2026) · International Conference on Consumer Electronics · doi:10.1109/icce67443.2026.11449902
“In the proposed DMS design, firstly a NIR-based video stream is inputted into the YOLOv9 [2, 3] based models to perform the object detection, head pose estimation, and gaze estimation.”
Autonomous Driving Robot Car Using External NPU for Object Detection
Kojima, A. (2026) · International Conference on Consumer Electronics · doi:10.1109/icce67443.2026.11449726
“From the original YOLOv3[5], many derivative versions have been developed, including YOLOv4[6], YOLOv5, YOLOv6[7], YOLOv7[8], YOLOv8[9], YOLOv9[10], YOLOv10[11], YOLO11, and YOLOv12[12].”
GGA: Gradient-Guided Augmentation for Robust Object Detection under Domain Shift
Lee, E., Yoon, J., Baek, M., & Paik, J. (2026) · International Conference on Consumer Electronics · doi:10.1109/icce67443.2026.11449691
“Deep learning-based models, particularly the R-CNN series [1] and YOLO [2], have demonstrated outstanding performance under ideal conditions, such as clear weather.”
MapSAM2: Adapting SAM2 for Automatic Segmentation of Historical Map Images and Time Series
Xia, X., Balestriero, R., Zhang, T., Zhou, Y., Ding, A., Saini, D., & Hurni, L. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3678420 · Cited by 2
“For historical map time series, we integrate a YOLO detector [39] to provide automatic prompts.”
Zhang, H., Tang, N., & Sun, J. (2026) · Scientific Reports · doi:10.1038/s41598-026-45851-0
“Single-stage detectors such as YOLO 20 and SSD 21 treat detection as a direct regression problem for higher throughput; the multi-box loss in SSD is formulated as: where L conf is the confidence loss, L loc is the localization loss, N is the number of matched boxes, and α balances the two terms.”
InfluentialJailani, B. M., Mustafa, M. F., Hamzah, A., et al. (2025) · doi:10.1109/icoias69065.2025.00011
“The introduction of the YOLO framework redefined real-time object detection by combining localization and classification into a single-stage network [2].”
Research and Optimization of Methods for Detecting Objects in Images
Podoliak, B., & Filimonova, T. O. (2026) · Artificial Intelligence · doi:10.15407/jai2026.01.049
Explainable hybrid AI CAD framework for advanced prediction of steel surface defects
Moon, C., Al–antari, M. A., & Gu, Y. H. (2026) · Scientific Reports · doi:10.1038/s41598-025-34320-9 · Cited by 1
Scratcher: An automated machine-vision tool for dissecting the neural basis of itch
Nandi, D., Ravi, R. K., Prajapati, J. N., & Barik, A. (2026) · Neuroscience · doi:10.1016/j.neuroscience.2026.03.046 · Cited by 1
Smart Public Queue Traffic Analyzer and Decision Assistant Using Computer Vision
Priyadharshini, S. A. S. V., & R. Gopika Sri, S. D. K. (2026) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-32268
Benchmarking YOLO Variants on an Unseen Video: A Comparison of Inference Speed, GFLOPs, and Recall
Vibha, T. G., Theodore Chandra, S., & Sivaramakrishnan, S. (2026) · International Research Journal of Multidisciplinary Technovation · doi:10.54392/irjmt26212 · Cited by 1
Yang, Y., Ding, B., Yang, G., et al. (2026) · Energy Technology · doi:10.1002/ente.202501621
Diamond-DETR: lightweight real-time quality evaluation algorithm for synthetic diamonds
Yan, X., Yang, S., Zhang, S., Li, X., & Li, A. (2026) · Scientific Reports · doi:10.1038/s41598-026-44951-1
Unveiling the Power of YOLO: Enhanced Number Plate Recognition System
Srivastava, B., & Chahal, P. (2025) · National Conference on Electrical and Electronics Engineering (ELECO) · doi:10.1109/ice367573.2025.11448829
“Although faster than traditional methods like Faster R-CNN, YOLOv1 had lower accuracy, especially for small objects, achieving around 45 FPS with a CNN architecture comprising 24 convolutional and 2 fully connected layers [12].”
YOLO-based Lightweight Drill Detection for Coal Mines
Zhang, D. (2026) · Journal of Computer Science and Artificial Intelligence · doi:10.54097/zfzay098
“At present, mainstream object detection models represented by YOLO series [3], Faster R-CNN [4] and CenterNet [5] have achieved excellent performance in general scenarios.”
Automated FHWA Vehicle Classification System via LiDAR and Camera Data Fusion
Arthur, E., Zhang, L., & Adu-Gyamfi, Y. O. (2026) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981261429868
Zhang, X.-W., Zeng, H.-R., Zhang, X.-Q., et al. (2026) · IEEE transactions on multimedia · doi:10.1109/tmm.2026.3678441
Semi-Supervised Learning Method for Drill Pipe Detection
Zhang, D. (2026) · Academic Journal of Applied Sciences · doi:10.54097/vz9ddf87
“One-stage detection algorithms represented by the YOLO[1] series have been widely applied in underground equipment monitoring tasks due to their high detection accuracy and real-time performance.”
Abella, M. P. C., & Catedrilla, G. M. B. (2025) · doi:10.1109/icaisg68699.2025.11452147
Liu, H., Wang, L., Zheng, Q., Zhao, G., & Xu, H. (2026) · Symmetry · doi:10.3390/sym18040587 · Cited by 2
YCL Based Smart Glasses for Visually Impaired
Wong, W. S., Loh, Y. W., Lee, W. X., et al. (2026) · Journal of Engineering Technology and Applied Physics · doi:10.33093/jetap.2026.8.1.18
ACA-Net: A Modified YOLOv11 Network for Accurate Wildlife Detection
Yu, S., Ren, X., & Sun, Q. (2025) · doi:10.1109/icaisg68699.2025.11452137
“In general, deep learning-based object detection could be divided into two-stage (such as Faster R-CNN [1]) and single-stage approaches (such as YOLO [2]).”
Deep reinforcement learning based dynamic printing control in Large Format Additive Manufacturing
Xie, H., Hoskins, D., Rowe, K. G., & Ju, F. (2026) · IISE Transactions · doi:10.1080/24725854.2026.2649852 · Cited by 1
NF-DETR: A real-time detection method for military vessels and onboard small targets
Liu, H., Xu, J., Lin, Z., et al. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08477-7
Cao, T., Li, W., Xu, H., et al. (2026) · Journal of Grid Computing · doi:10.1007/s10723-026-09828-7
“In contrast, one-stage methods such as YOLO [9] offer greater efficiency but perform worse on small targets.”
Photo Captioning for Assisting Blind and Visually Impaired People Using Gated Attention
Pavithra, K., Praveena, K., Padmavathy, S., Ishwarya, A., & Velmurugan, M. (2026) · International Conference on Computing, Communication, Control and Cyber-Physical Systems (I5CPS) · doi:10.1109/i5cps67958.2026.11452630
“Redmon et al. introduced the first YOLO model, which revolutionized object detection as a single regression task [13].”
A CNN–LSTM Framework for Player-Specific Baseball Pitch Type Prediction from Video Sequences
Chang, C.-C., Wei, C.-H., Li, H.-C., & Hsiao, S. (2026) · Applied System Innovation · doi:10.3390/asi9040075 · Cited by 3
Multi-View Camera-Based UAV 3D Trajectory Reconstruction Using an Optical Imaging Geometric Model
Ji, C., Wang, Y., Yi, J., Zheng, X., Geng, W., & Cheng, L. (2026) · Electronics · doi:10.3390/electronics15071425
Moustafa, A., Khalid, H. Y. I., Najimaldeen, R. M. S., et al. (2026) · Zanin Journal of Science and Engineering · doi:10.64362/zjse.159
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.