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 45,901–46,000 of 51,399
Wang, L., Liu, Z., Lv, T., Wang, X., & Qiu, T. (2025) · Materials · doi:10.3390/ma18225159
“Redmon et al. redefined object detection as a regression problem through the YOLO algorithm, which completes detection in a single forward pass via a convolutional neural network, significantly simplifying the detection process and improving processing speed [30].”
YOLO-MaskRCNN: An Improved Statistical Model for Blood Cell Classification
王, 家. (2025) · Statistics and Applications · doi:10.12677/sa.2025.1411318
Li, Q., Zhou, Y., Zhang, J., Xu, L., Chen, J., & Li, Z. (2025) · Applied Sciences · doi:10.3390/app152212043 · Cited by 1
Zhang, X., Zheng, Q., Zhu, P., & Ma, W. (2025) · Applied Sciences · doi:10.3390/app152212033 · Cited by 1
Xingxing, 陈. C., Shenglin, 李. L., & Xiangling, 周. Z. (2025) · Laser & Optoelectronics Progress · doi:10.3788/lop251450
STARS: Semantics-Aware Text-guided Aerial Image Refinement and Synthesis
Townsell, D., Chen, L., Xie, M., Pan, C., & Zhang, W. (2025) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2025.104561
Characterizing manufactured home parks in North Carolina: A computer vision based approach
Khanal, K., Kaza, N., Hino, M., & Sebastian, A. (2025) · Environment and Planning B Urban Analytics and City Science · doi:10.1177/23998083251395471
“…detector that detects an object of interest in an image with a single pass, in contrast to classical two-stage object detectors in which regions of interest are fi rst identi fi ed and later classi fi ed, allowing faster inference, at the expense of accuracy for smaller objects (Redmon et al., 2016).”
Han, S., Zhang, T., Yi, Z., Jiang, W., Liu, B., He, J., & Li, L. (2025) · Medical Physics · doi:10.1002/mp.70130
“11,12 Wang et al. 13 employed YOLOv5m for ACL detection in each slice, followed by two classifiers for three-class classi-fication, while Tran et al. 14 used a standard CNN for localization.”
Huang, J., Tang, Y., Wang, K., et al. (2025) · Journal of Forensic Sciences · doi:10.1111/1556-4029.70216
“Since single-target detection accuracy is higher than multi-target detection accuracy [ 27 ], YOLOv11 was utilized to train each type of minutiae of fingerprints individually.”
Car Number Plate Recognition Using Image Processing
Bidkar, P. M. V., Bagal, M. S., Chavan, M. A., et al. (2025) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-29492
An Improved CenterNet2 Model for Long-Arm Engineering Vehicle Object Detection
Li, Y., Hua, J., Wang, Z., Li, X., & Tian, F. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229314
“There also appear numerous typical networks such as SSD [6], Yolo series [7-11], and RetinaNet [12].”
MPFDAN: Multi-Perspective Feature Dynamic Adaptation Network for Domain Adaptive Object Detection
Yang, H., Liu, K., Weng, W., Dai, W., Cichocki, A. S., & Kong, W. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229255 · Cited by 1
EduSCARA: An Open Source RRPR Educational SCARA PlatformPreprint
Clark, A., Martínez-Hernández, U., & Assaf, T. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5749008
Learning Location-Aware Visual Representations Through Anchor-Based Contrastive Learning
Kumar, C., Herrera-Gerena, J., Just, J., Darr, M. J., & Jannesari, A. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-07109-5_18
Ethical Considerations in AI-Based Brain Tumour Diagnosis
Rangelov, D., Miltchev, R. I., & Genchev, E. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-07109-5_5
Xu, Z., Mazurkiewicz, L. E., Olivetta, M., et al. (2025) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2025.11.12.688106
Lin, M. (2025) · International Conference on Computers, Information Processing and Advanced Education (CIPAE) · doi:10.1109/cipae66821.2025.00048
“One-stage methods (e.g., YOLO [5], FCOS [6], ATSS [7]) remove the RoI step and directly predict bounding boxes and classes.”
Ivanova, E. I., & Gerov, D. S. (2025) · International Scientific Conference on Computer Science (COMSCI) · doi:10.1109/comsci67172.2025.11224987
“Previous research has extensively explored object detection models like YOLO [4], SSD [5], and Faster R-CNN [6].”
Recognition of Handwritten Characters using Deep Learning Algorithms
Bennet, M. A., Remya, R., Selvakani, R., Joel, T., Sindhu, U. L., & Raja, C. (2025) · IEEE International Conference on Compute, Control, Network & Photonics (ICCCNP) · doi:10.1109/icccnp63914.2025.11233751
MVBMR: Multi-View Breast Mass Recognition Network
Yuan, H., Zhao, N., Dong, S., & Yan, Y. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227712
CharDetNet: Lightweight Ancient Character Detection with Mixed Attention
Song, Y., Zhang, C., & Wu, M. (2025) · doi:10.1109/ijcnn64981.2025.11228720
Chan, S., Chen, X., Wang, C., Wang, W., Mao, J., & Hu, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227288
“Some approaches utilize more refined detectors [1]–[4] and appearance feature extractors [5]–[9] to obtain the location and appearance embedding vectors of target vehicles.”
Towards Reliable X-ray Security: SCE-DETR for Detection and Occlusion Handling
Wang, Z., Fan, Y., & Ye, B. (2025) · doi:10.1109/ijcnn64981.2025.11227877
“Convolutional neural networks (CNNs) have significantly accelerated progress in this field [3], with the YOLO series models, from YOLOv1 [4] to YOLOv10 [5]—introducing continuous innovations to improve speed and accuracy, making them indispensable in object detection.”
Yin, X., Yuan, W., Wu, X., Fu, L., & Huang, D. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227672 · Cited by 1
“In contrast, one-stage detec-tors—like the YOLO family [7]–[12], SSD [13], Retina-Net [14], and CenterNet++ [15]—frame detection as a regression *Corresponding author: wuxc3@stu.xjtu.edu.cn problem, achieving an optimized equilibrium between computational efficiency and detection accuracy.”
Research on a PCB Defect Detection Algorithm Based on Improved YOLOv8n
Guo, P., & Chen, Y. (2025) · International Conference on Electronic Information Technology (EIT) · doi:10.1109/eit67313.2025.11231915
“Single-stage algorithms, such as YOLO [5] and SSD [6] , perform classification and localization together, providing fast detection.”
Automatic Detection of Airway Mucus Plugs in CT Scans Using BiVariant Attention and ROI Masking
Wang, Y., Liu, K., Zhang, Y., Niu, K., Chang, C. H., & He, Z. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228814
LJ-DETR: LightWeight RT-DETR Algorithm with Joint Encoder for Steel Surface Defect Detection
Wang, C., Wang, T., Li, H., Wang, M., Xia, J., & Liu, X. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228421
“1) yolov10: In object detection tasks, the YOLO (You Only Look Once) series [15] has been at the forefront of real-time detection methods.”
Study of YOLOv8-Based Techniques for Detecting Road Surface Damage
Yin, C., Yu, J., & Li, S. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228986
“However, YOLOv1 struggles with detecting small targets [10].”
MCRO-YOLO: An evolved Version of YOLO for Small Object Detection
Li, J., Shi, Q., Yang, F. O., Fu, S., & Hou, F. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227312 · Cited by 2
“YOLO [13] pioneered single-stage detection by framing object detection as a regression problem, making global predictions across the entire image, and achieving both high speed and robust generalization.”
Detecting Large Aspect Ratio Objects with Geometry-Aware in Remote Sensing Images
Song, P., Yang, Y., & Wang, C. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229040
“Early classical methods include R-CNN series [1], [2], YOLO series [3], [4], etc.”
Traffic Sign Small Object Detection Algorithm Based on Lightweight Structure Design
Chen, S., An, P., Zeng, S., Lin, C., Wu, Z., & Ke, Z. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228472
“In contrast, single-stage detection algorithms, represented by the You Only Look Once (YOLO) series [10-20], cover the entire image pixel by pixel by using candidate boxes of different sizes.”
STPYOLO: A Lightweight and Efficient Model for Small Target Pest Detection
Wang, J., Jia, Z., Zhou, G., et al. (2025) · doi:10.1109/ijcnn64981.2025.11227202
“One-stage object detection models for pest detection include YOLO [2] and other similar models.”
Zheng, X., & Shao, B. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228159
“…[13] and proposal generation [14], but suffer from slow speed and missed detections; single-stage detectors such as SSD [15] and YOLO series [16, 17, 18] improved speed and recall rates through enhanced feature pyramids [19, 20] and backbone networks [21, 22], but still have limitations in…”
Yu, Y., Bao, R., Gu, J., et al. (2025) · doi:10.1109/ijcnn64981.2025.11229226
Bao, Z., & Ibrayim, M. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229105
“Text detection extends general object detection methods like Faster R-CNN [3], YOLO [4], and SSD [5].”
ASTD-ABC: Arbitrary Shape Text Detection with Adaptive Points and B-Spline Curves
Liang, X., Wang, H., Dan, Y., Han, J., Zhao, B., & Zhang, Q. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228698
AI-Camera-Based Indoor Positioning System for the Wireless-Powered Internet of Things
Dilazapira, Mubarok, S., & Aziz, A. A. (2025) · IEEE International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA) · doi:10.1109/icsima66552.2025.11233335
MAF-DETR: Defect Detection with Multi-Scale Adaptive Feature Fusion
Zhang, C., Wu, W., Niu, H., et al. (2025) · doi:10.1109/ijcnn64981.2025.11227633
“FollowingR-CNN [6], object detection models like Faster R-CNN [7], YOLO [8], and RT-DETR [9] have advanced defect detection.”
NR-DETR: A Lightweight Real-Time Fabric Defect Detection Model with Multi-Scale Enhancement
Hao, H., Xu, L., Gao, C., et al. (2025) · doi:10.1109/ijcnn64981.2025.11229106
Omalla, M. K. T., Abdullah, H., & Ahmad, N. B. (2025) · IEEE International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA) · doi:10.1109/icsima66552.2025.11233353
“YOLO operates as a single-stage detector that simultaneously predicts object classes and bounding box coordinates from input images in a single forward pass, making it highly suitable for real-time applications such as intelligent transportation systems (ITS) [2, 4, 5].”
DyFusion-YOLO: An Enhanced Model for Small Object Detection in Aerial Imagery
Wang, X., Wang, Y., Liu, T., et al. (2025) · doi:10.1109/ijcnn64981.2025.11228116
“As deep learning methods continue to evolve, especially single-stage detectors (e.g., the YOLO [1] family, which stands for “You Only Look Once”) have shown great potential for object detection.”
A Study on Road Damage Detection Methods Based on YOLOv11
Yin, C., & Yu, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229375
“YOLO treats detection as a regression problem, predicting multiple bounding boxes and categories simultaneously in a single pass, which significantly speeds up the detection process [6].”
Enhanced YOLOv7 for UAV Platforms: Achieving High Precision with Low Resource Utilization
Sui, C., Zhang, J., Jin, L., Lu, Q., & Chen, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228323
“In contrast, single-stage algorithms, such as the YOLO [5] series, utilize convolutional neural networks to perform both object localization and classification simultaneously.”
Real-Time Small Object Detection System for Herbal Pieces Using YOLO with a Multi-Cluster Strategy
Wang, Y., Yu, X., Wang, F., et al. (2025) · International Conference on Platform Technology and Service · doi:10.1109/platcon68373.2025.11239204
“Introduced by Joseph Redmon and his team in 2016 [5] , the YOLO model differs from traditional methods by treating object detection as a unified regression problem.”
HSDNet: Hybrid Super-Resolution Assisted Detection Network for Surface Defect Detection
Lin, X., Yang, X., Sun, Y., et al. (2025) · doi:10.1109/ijcnn64981.2025.11227791
“In the era of deep learning, numerous deep-learning based object detectors including [1] series and [2] series achieve outstanding performance in various scenes.”
A Real-time Small Object Detection Model for UAV Aerial Images in Renewable Energy and Power Systems
Li, T., Zhang, C., Lian, X., Liu, Z., & Li, Y. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229294
“Notable examples of these methods include SSD [11], FSSD [12], and the YOLO series [13]–[19], among others.”
Chain of Thoughts Based Prompting for Incident Detection Using Large Language Model
Chambers, L., Udokisai, A., Mahmoud, H., Ismail, K. J. A., Mohammed, N., Cervantes, W., Gaber, M. M., & Bhana, R. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229199
“Most notably, the YOLO [19] and Faster Region Convolutional Neural Network (Faster R-CNN) [20] algorithms [21] are leading the way.”
DDYOLO: Efficient Rainy Scene Object Detection with Deraining Guidance
Chen, X., Yan, T., Wang, C., Liu, J., Zhang, Z., & Huang, W. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228916
“Without an explicit region proposal step, these methods are much faster than the former, with the YOLO series [4], [16] and SSD [2] being typical examples, well-suited for real-time processing scenarios.”
YOLO-FEE: An Improved Fabric Defect Detection Model Based on YOLOv11s
Wen, Z., Pan, W., Dong, Z., Wang, J., Su, J., Guo, F., Wu, X., & Zhou, Y. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229236 · Cited by 1
“With the continuous development of the YOLO series, the detection accuracy and speed have been significantly improved, especially with YOLOv4 [4] and YOLOv5, they have more efficient network architectures and optimization strategies, which have been widely applied in textile defect detection.”
Mignot, T., Ponchon, F., Derville, A., Duffner, S., & García, C. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228587 · Cited by 1
“The YOLO (You Only Look at Once) architecture [3] is a well-known framework that has been extensively investigated in this domain, with a comprehensive review provided in [4].”
YOLO-GSW: A Lightweight Multi-Granularity Fruit Detection Algorithm for Complex Life Scenarios
He, Y., & Yang, H. (2025) · doi:10.1109/ijcnn64981.2025.11229111
“Typical examples include the R-CNN [1], Fast-RCNN [2], and Faster-RCNN [3] proposed by Girshick et al. The other is the single-stage, where YOLO [4]–[10] is most commonly used.”
Multimodal Object Detection by Adaptive Channel Enhancement and Attention Fusion
Mei, Y., Zhang, T., & Tan, H. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227959
“…Neural Networks (CNNs), researchers began exploring the application of CNNs in object detection, leading to the development of deep learning-based frameworks, including two-stage methods (R-CNN [19], Fast R-CNN [20], Faster R-CNN [21]) and one-stage methods (SSD [22], YOLO series [23, 24, 25]).”
YOLO-fungi: Detection Method for the Growth Stages of Edible Fungi under Complex Environments
Zhang, W., Wang, X., Huang, J., & Yang, H. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227700 · Cited by 1
“The one-stage detection algorithm includes SSD [12], YOLO [13], etc., which discards region proposal generation, instead directly transforming object localization into a regression problem, to predict the position coordinates and class probabilities of objects for faster detection.”
Marceddu, A. C., Dilillo, N., Sergio, L. D., Ruiu, P., Lagorio, A., Casu, F., Grosso, E., Ferrero, R., & Montrucchio, B. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228494 · Cited by 1
“Second, labels in YOLO [5], PascalVOC [6], and COCO [7] formats have been added to enable both image classification and object detection tasks.”
Para-X: Graph-based Facial Paralysis Detection using Structural Deformations of Facial Expression
Sharma, N., Singh, K., & Singh, D. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228302 · Cited by 4
From Perception to Distance: A Lightweight RGB-D Pipeline for Adaptive Cruise Control
Fathy, Y. M., Ibrahim, H. H., & Shehata, O. M. (2025) · Novel Intelligent and Leading Emerging Sciences Conference (NILES) · doi:10.1109/niles68063.2025.11231963
“Various detection models have been developed over the years, including YOLO [6], Faster R-CNN [7], and EfficientDet [9], each offering trade-offs between accuracy, speed, and computational efficiency.”
A Robust Multimodal Model with Context-Rich Prompts for Traffic Light State Detection
Zheng, J., Murphey, Y. L., Giordani, B., Persad, C. C., & Maher, A. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228651
“Typical models include You Only Look Once (YOLO) [8] and Single Shot multibox Detector (SSD) [9].”
FLEX-YOLO: A Feedback-Driven Lightweight and Enhanced Exchange Network for UAV Aerial Images
Ding, N., Li, X., Ma, J., Huan, R., & Zhao, X. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227786
“In contrast, one-stage methods, like the YOLO series [7]–[10], simplify detection into a single regression task, enabling efficient real-time detection and better performance on small objects.”
DMP-DETR: Dilation Multi-Scale and Partial Channel Gated Fusion Attention For Steel Defect Detection
Zhao, S., Wu, C., & Yu, F. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229140
“A prominent example of one-stage detectors is You Only Look Once (YOLO) [6], which divides an image into a grid and predicts bounding boxes and class probabilities within each cell.”
Multiscale Attention Knowledge Distillation for Object Detection
Zhang, F. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227248 · Cited by 1
“However, high demands for lower latency bring one-stage detectors [34], [37] under the spotlight, which achieve classification and location of targets through the feature map directly.”
DH-YOLO: Double-Head YOLO for Detecting Refrigerant Leakage Smoke in Air Conditioners
Jiang, Y., Wang, R., Hua, J., et al. (2025) · doi:10.1109/ijcnn64981.2025.11227583
Pay, Q. B., Baskaran, V. M., Loo, J. Y., et al. (2025) · doi:10.1109/ijcnn64981.2025.11229177
“Although popular fields such as object detection [1], [2] identify entities and action recognition [3], [4] discern specific human actions, they often present a fragmented view of the visual narrative.”
Enhanced YOLOv8 for Road Defect Detection in Complex Scenarios
Hu, J. P., & Jia, Y. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228455
“Representative algorithms include YOLO [6], RetinaNet [7], and SSD [8], among others.”
Enhance Object Detection via Instance Transformation Between Categories Using Diffusion Models
Ma, H., & Ren, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227834
“Automated detection using deep learning models like Faster R-CNN [1], YOLO [2], and DINO [3] has shown significant progress.”
LW-DETR: Vision Transformer-Based Network for Transmission Line Fittings Defect Detection
Liu, D., & Liu, P. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229094
“In contrast, one-stage approaches like SSD [4] and YOLO [5] directly predict object categories and bounding boxes, bypassing the candidate region generation step, which simplifies the process and improves efficiency.”
Lightweight Object Detection via Joint Region-Context Feature Representation
Li, C., & Zhang, T. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228417
“In contrast, one-stage networks like YOLO [5], [13]–[18]and SSD [6] perform detection in a single step, making them faster but often less accurate than two-stage networks.”
DQ3D: Depth-guided Query for Transformer-Based 3D Object Detection in Traffic Scenarios
Wang, Z., Li, W., & Wu, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227159 · Cited by 1
“Object detection has been a central research topic in computer vision for decades, with traditional approaches [12]– [16] making significant strides in recent years.”
Gunawan, T. S., Azlin, A. A., Kartiwi, M., et al. (2025) · IEEE International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA) · doi:10.1109/icsima66552.2025.11233477
“YOLOv12, the latest version, incorporates attention-based mechanisms and architectural improvements that enhance detection speed and efficiency for embedded systems [10, 11].”
Contextual Self-Learning Approach for Continuous Object Detections in Expansive Visual Domains
Bharatia, D. V., & Bharatia, H. V. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227623
Maximizing Eye Segmentation Accuracy with YOLO
Vecchi, A. D., Azzi, S., Bartoli, P., et al. (2025) · doi:10.1109/ijcnn64981.2025.11228981
“In this paper, we focus on the usage of state-of-the-art segmentation models based on CNNs, more specifically You Only Look Once (YOLO) [9], to segment the eye in its three main regions, i.e. sclera, iris, and pupil.”
Masked strategies for images with small objects
Gillis, H. M., Hill, M., Hollensen, P., et al. (2025) · doi:10.1109/ijcnn64981.2025.11227847
“The authors then applied scaled-down networks derived from the You Only Look Once (YOLO) [13] and ResNet-18 [14] neural networks.”
NP4Q: Nice Point for Post-training Quantization of Object Detection Models
Zhou, S., Jiang, J., Xu, J., Li, L., & Zhu, M. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228181
Detection of Small UAV Objects with Density Information
He, F., Yao, W., Jiang, X., & Liu, B. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11229160
Ding, J., & Gu, J. (2025) · doi:10.1109/ijcnn64981.2025.11227335
“With the continuous development of deep learning, modern target detection methods for remote sensing images have increasingly used neural networks, especially convolutional neural networks [6] [7].”
Prior-Guided Feature Fusion Network for Driving Obstacle Detection in Foggy Weather
Zhang, D., Geng, X., & He, M. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227356
“One-stage target detection algorithms perform classification and position regression directly on input images without generating candidate boxes, exemplified by the YOLO series [7] and SSD [8].”
Shubqizam, N. N. N. M., Zulkurnain, N. F., Gunawan, T. S., et al. (2025) · IEEE International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA) · doi:10.1109/icsima66552.2025.11233761
A Weighting Loss Approach for Transformer-Based Object Detection
Tzimas, M. D., Mygdalis, V., & Pitas, I. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227605
“The YOLO (You Only Look Once) family [6]–[8], [23], represents a groundbreaking series of CNN-based real-time object detectors, characterized by either anchor-based or anchor-free architectures.”
LOVD: Open-Vocabulary Object Detection Based on LLM-Enhanced Agents
Li, X., & Xu, J. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11227349
“With the development of deep neural networks, numerous studies [5] [6] [7] [8] have achieved significant breakthroughs in object detection.”
HR-OTTA: Robust Online Test-Time Adaptation with Hybrid Fine-Tuning
Lei, J., & Pernkopf, F. (2025) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn64981.2025.11228164
“The results in this paper are based on evaluations conducted on three corruption datasets, meaning the findings may not fully generalize to TTA approaches for other tasks, such as semantic segmentation [3], object detection [4], or natural language processing [45].”
Optimized YOLO11 for Accurate Wind Turbine Blade Defect Detection
Liu, C.-X., Liu, N., Zhuang, X., et al. (2025) · IEEE International Joint Conference on Neural Network · doi:10.1109/ijcnn64981.2025.11227383
“Among these emerging approaches, single-stage detection models, particularly the YOLO series [15] [16] [17] [23] [20], have become the ideal choice for real-time defect detection because of their excellent balance between detection speed and accuracy.”
Zhao, A., Yin, J., Wang, W., Guo, Z., & Zhu, L. (2025) · Electronics · doi:10.3390/electronics14132702
Real Time Object Detection Tracking Using YOLO and Deep Sort
Hemashree, C., S, P. B., Pruthvi, H. N., et al. (2025) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-29912
A Task-Specific Feature Fusion Strategy For Small Object Detection
Shang, Z., Zhang, Z., & Zhao, M. (2025) · Lecture notes in computer science · doi:10.1007/978-981-95-3001-4_8
叶, 阳. (2025) · Artificial Intelligence and Robotics Research · doi:10.12677/airr.2025.146134
Enhancing Construction Site Safety and Efficiency with YOLO v8-Based Computer Vision Model
Sabek, M., González, V. A., Mei, Q., & Lee, G. (2025) · Lecture notes in civil engineering · doi:10.1007/978-3-031-97697-1_21
Real-Time Obstacle Avoidance Using Monocular Depth Estimation and Optical Flow Techniques
Shaikh, M. A., Syed, M. A., Yadav, C., et al. (2025) · Smart innovation, systems and technologies · doi:10.1007/978-981-96-6250-0_40
Menyhárt, T., & Lakatos, R. (2025) · Future Transportation · doi:10.3390/futuretransp5040175
“Faster R-CNN (Region-based Convolutional Neural Network [1], YOLO (You Only Look Once) [2], and Vision Transformer [3] are just a few examples of these complex architectures.”
Vision-based human sensor networks for wildfires
Lever, J., Schneider, R., & Arcucci, R. (2026) · Natural Hazards · doi:10.1007/s11069-026-08240-4
“YOLO models represent a class of single-stage detectors that process images by dividing them into grid cells during convolution, enabling real-time performance while maintaining competitive accuracy (Redmon et al. 2016 ).”
Comparing YOLO Models for Real-Time ASL Recognition
Gomez, A., & Arzuaga, E. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-032-09694-4_17
Ship Detection in SAR Images Using DETR with Feature Pyramid Network
Wang, C., Gong, J., & Guo, B. (2025) · Smart innovation, systems and technologies · doi:10.1007/978-981-96-4434-6_31
Deep Learning Application for Egg Quality Detection in Ephestia Kuehniella
He, D.-W., & She, C. (2025) · Smart innovation, systems and technologies · doi:10.1007/978-981-96-4434-6_3
Enhancing POI Recognition with Micro-Level Tagging and Deep Learning
Messios, P., Dionysiou, I., & Gjermundrød, H. (2025) · Big Data and Cognitive Computing · doi:10.3390/bdcc9110293
Otel Endüstrisinde Dijital Dönüşüm: Konuk Deneyimini İyileştiren Teknolojik Çözümler
Çıtak, M. E., Hatun, H. İ., Ecevit, A., Körpe, B., & Uçar, M. (2025) · Selçuk Turizm ve Bilişim Araştırmaları Dergisi · doi:10.71276/stbad.1772965 · Cited by 1
SmartGuard: Advanced Security System with YOLOv8
Ch, V., Kakulapati, V., Reddy, P. D., et al. (2025) · Lecture notes in electrical engineering · doi:10.1007/978-981-95-0269-1_174
Hassan, I. M. E., Al-Mallahi, A. A., & Shirzadifar, A. (2025) · Smart Agricultural Technology · doi:10.1016/j.atech.2025.101637 · Cited by 1
ROOT: Region-Word Alignment With Partial Optimal Transport for Open-Vocabulary Object Detection
Deng, J., Lei, Y., Li, W., & Duan, L. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3627395
“Early convolutional neural networks (CNN) [6] based object detectors [1], [3], [25], [26], [27], [28], [29] can be summarized into two distinct directions, i.e. , two-stage or one-stage models.”
Gabr, R. H., Sharfo, S. M., Zekry, R. S., Alraee, A. O., Elghobashy, M. A., & Labib, F. E.-Z. M. (2025) · Biomedical Engineering Applications Basis and Communications · doi:10.4015/s1016237225500681
Lim, S., Kim, S., & Park, J. (2025) · Journal of Advanced Marine Engineering and Technology · doi:10.5916/jamet.2025.49.5.346
Deep Learning Models for Corn Kernel Counting and Yield EstimationPreprint
Ghimire, B., Lacerda, L. N., Bourlai, T., et al. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5768641
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.