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 50,001–50,100 of 51,398
Sabouri-Ghomi, M., & Bhowmick, A. K. (2026) · Numerical Methods in Civil Engineering · doi:10.66224/nmce.2511.1109
Truşcǎ, M. M., Depauw, M., Fantoli, M., Soen, V., Daele, I. D., Verbruggen, K., & de Cruys, T. V. (2026) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-026-00605-6
Chen, Y.-K., Su, Z.-S., Tsai, Y.-L. S., Chang, S.-W., Han, J.-Y., & Huang, C.-J. (2026) · International Journal of Remote Sensing · doi:10.1080/01431161.2026.2705396
Pandey, P., & Mandal, S. (2026) · Multimedia Tools and Applications · doi:10.1007/s11042-026-21668-x
Geometry-Guided Affine-Invariant ROI Extraction for Palm Biometrics
Nanda, V., Pillai, M., Joseph, A., et al. (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619844
“Architectures such as YOLOv3-tiny and YOLOv5-lite [17], which are derived from efficient one-stage detectors within the YOLO (You Only Look Once) framework, have been widely used for detection-based ROI extraction [14].”
3D Human Pose Estimation from Monocular Video Sequences in Underwater Scenarios
Liang, S., Liu, H., Liu, P., Zhang, D., Yu, R., Zhou, X., & Zhou, Z. (2026) · Sensors · doi:10.3390/s26154738 · Cited by 1
Deep learning–based real-time droplet size analysis for polystyrene microsphere synthesis
Dong, Z., Xiao, Y., Wu, S., Wang, L., Ma, S., Li, Z. Y., Wang, Z., Zhang, Q., & Yang, C. (2026) · Chemical Engineering Science · doi:10.1016/j.ces.2026.124788 · Cited by 1
Deep Learning-Based System for Pineapple Eye Scanning and Extraction Using YOLOv8n and GRBL
Franco, R., Gonzales, J., Nieves, A., Carrizales, J., Figueroa, J., & Malpartida, E. A. S. (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619721
“Future work includes multi-view scanning, improved calibration routines, more extensive datasets, and robustness strategies commonly used in detection systems [7], [8].”
Toderici, D., Rodanciuc, T.-G., Micu, G.-A., Rughinis, R. V., Lupşa, S.-R., & Țurcanu, D. (2026) · Electronics · doi:10.3390/electronics15153295
A Lightweight Feature Transformation Network with Progressive Training for Rainy Object Detection
Wang, T., & Wei, J. (2026) · Journal of Computing and Electronic Information Management · doi:10.54097/v3s89j21
“In recent years, deep learning-based detectors (e.g., two-stage Faster R-CNN[1], one-stage FCOS[2], and the YOLO series[3-9]) have achieved remarkable performance on standard academic benchmarks such as MS-COCO[10] and PASCAL VOC[11].”
Zhang, L., & Xiong, Z. (2026) · Algorithms · doi:10.3390/a19080620
Wang, P., & Qian, K. (2026) · doi:10.1109/cicse70286.2026.11621017
“The single-stage end-to-end design addresses the trade-off between localisation accuracy and inference speed [9]-[17].”
Enhancing VGGT-Based 3D Scene Reconstruction via Object Segmentation
Zhang, X., Shi, H., Wang, X., Li, B., & Cai, Z. (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619678
Reyes, G., Tolozano-Benites, R., Villamar, A. C., et al. (2026) · Applied Sciences · doi:10.3390/app16178685
Scalable Deep Learning for Intelligent Weld Quality Assessment Toward Structural Failure Prevention
Sabri, K. (2026) · Journal of Failure Analysis and Prevention · doi:10.1007/s11668-026-02516-4 · Cited by 1
“[33]: while the classification loss is expressed as As illustrated in Fig.”
SOEM-DETR: an efficient real-time drone infrared image photovoltaic defect detection model
Peng, L., Zheng, S.-K., & Wang, D.-L. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01949-2
“These methods include single-stage approaches (such as Single Shot MultiObject Detector SSD [11] and the YOLO series [12]) and two-stage approaches (such as the Region-based Convolutional Neural Network RCNN series [13]).”
Liu, Y., Gao, P., Ying, K., Wu, B., Xiao, D., Qiu, Y., Feng, D., He, R., He, D., Gui, L., et al. (2026) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2026.3717119
Research of AL-YOLOv8 for Chip Surface Defect Detection
Yuan, S., & 尹文露 (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619634
“Conversely, one-stage models excel in detection speed, exemplified by the YOLO [5] series.”
Sun, J., Xie, Z., & Li, Y. (2026) · Buildings · doi:10.3390/buildings16152975
Lightweight vision transformer for adversarially robust object detection in autonomous vehicles
Okul, Ş. (2026) · Gümüşhane Üniversitesi Fen Bilimleri Enstitüsü Dergisi · doi:10.17714/gumusfenbil.1866586
BiDetTrack: Enhancing MOT via Dual-Branch Candidate Generation and Spatio-Temporal Query Matching
Jiang, X., Zhu, Y., Zong, P., Song, M., & Ou, Z. (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619910
“Early systems like SORT [12] and DeepSORT [13] established the tracking-by-detection paradigm, employing detectors such as YOLO [16] or YOLOX [18] followed by Kalman filtering [19] and Hungarian algo-rithm [20] for association.”
Sanchez, G. N. V., Bailon, J. A. M., Espinoza, C. H. I., et al. (2026) · International Conference on Image and Graphics Processing (ICIGP) · doi:10.1109/icigp68997.2026.11619820
“The use of YOLOv5, YOLOv8 and YOLOv11 enables simultaneous handling of classification and detection tasks, while ResNet provides a robust alternative for categorization based on visual attributes [15], [16].”
LCA-Net: A Lightweight Network for Small Object Detection in Road Traffic Scenes
Lin, S., Yun, B., Lin, Z., Shen, J., & Xu, Q. (2026) · Information · doi:10.3390/info17080724
A real-time multi-security adaptive HEVC selective encryption scheme for traffic surveillance
Cai, Z., Sheng, Q., He, Y., Fu, C., Liu, X., & Wen, J. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01945-6
“Section2 describes some background knowledge on hyperchaotic Lü system and YOLO (You Only Look Once [33])-based object detection algorithms.”
Guimaraynz, H. D., Tashtoush, T., Oliva, D., et al. (2026) · Measurement Science and Technology · doi:10.1088/1361-6501/ae90d1
“Early systems relied on general-purpose object detectors such as Faster R-CNN [34], SSD [35], and the YOLO family [36], which proved robust and efficient across a wide range of urban scenarios.”
Adapted RD-YOLO-Based Defect Detection for Power Electronic Equipment
Chu, H., Zhang, Z., Wang, Q., et al. (2026) · Applied Sciences · doi:10.3390/app16157480
Yu, J., Peng, Y., Li, R. K., Shi, F., Chang, X., Chen, Y., Fang, J., Li, M., & Wei, D. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3715307 · Cited by 1
“Two-stage methods, such as Faster R-CNN [11], exhibit strong region proposal modeling capability, whereas one-stage detectors represented by the YOLO series [12] have been widely used in real-time object detection tasks due to their compact architectures and fast inference speed.”
Research on Visual Impairment Assistive Interactive System Based on Indoor Simulation Scene
Zhou, X., Ren, J., Zhou, J., & Zhou, X. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621447
“In recent years, target detection algorithms represented by YOLO (You Only Look Once) have made significant improvements in real-time detection speed and recognition accuracy [7], and have been gradually applied to fields such as intelligent monitoring, robot vision, and assistive interaction.”
Real-Time Pavement Defect Detection via Gated Multi-Scale Feature Fusion and Dynamic Activation
Ni, H., Wang, C., & Li, R. (2026) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs69761.2026.11613523
“First, conventional convolution operations, typically paired with static activation functions (e.g., SiLU or LeakyReLU), lack the nonlinear modeling capacity required to distinguish subtle or low-contrast defects (e.g., early-stage cracks) from noisy backgrounds [4].”
Liu, Y., Zhou, S., Li, C. Y., Chen, D., Liu, S., Zhang, Y., & Gao, Y. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3717755
Getz, M. P., Best, L. R., Ostroverkhova, O., Melathopoulos, A. P., & Warren, T. L. (2026) · Remote Sensing in Ecology and Conservation · doi:10.1002/rse2.70091
Ali, Y. A. M., Kaymih, N. A., & Alawad, N. A. (2026) · doi:10.1109/ais69919.2026.11621717
“For example, YOLO (You Only Look Once) detection systems achieve as low as 100 ms latency for detecting PPE violations [4], [5], [15].”
Ofem, O. A., Osahon, O., Obono, O., Essien, E. E., Njama-Abang, O., Ele, B., David, O. E., Paul, T. B., Edim, E. B., & Ashishie, D. U. (2026) · International Journal of the Physical Sciences · doi:10.5897/ijps2026.5120
Pest Detection Model Based on Multi-Scale and Reparameterizatior Feature Fusion
Chen, G., Le, W., Jia, Z., & Li, J. (2026) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs69761.2026.11613395
“Compared to second-stage network algorithms, classic first-stage SSD and YOLO models have obvious advantages [6] by unifying object classification and localization within a regression network architecture, offering faster detection speeds.”
Raoufi, B., Sharify, H., Ramezanee, M. M., et al. (2026) · doi:10.1109/iicai70155.2026.11620903
“YOLO [14] reframed detection as a single-shot regression problem and enabled real-time inference, SSD adopted multi-scale feature maps for dense predictions, RetinaNet introduced focal loss to overcome extreme foreground-background imbalance, and the transformer-based DETR family eliminated…”
Maskless Selective Object Removal via a Dual-Pipeline Framework with SAM–SDXL and LaMa
Park, S., Gong, M.-G., Park, S.-J., et al. (2026) · AI · doi:10.3390/ai7080284
Szlachta, A., Wnęk, J., Drzał, J., et al. (2026) · Electronics · doi:10.3390/electronics15153320
Albalooshi, F. A., & Qader, M. R. (2026) · Computation · doi:10.3390/computation14080168
“Architectures like ResNet [5], YOLO [6], and EfficientNet [7] have been adapted for PCB inspection, demonstrating significant improvements in accuracy over traditional methods.”
A Multi-Modal Approach for Hydrogen Leakage Monitoring based on Thermal Vision
Nguyen, H. V., Tran, C., Tran, H., et al. (2026) · doi:10.1145/3815970.3815978
“The YOLO family provides single-pass, real-time detection suitable for embedded systems [18].”
A Lightweight CNN-Based Digit Recognition System from Video Frames
Zhan, Q., Fan, Y., & Deng, Y. (2026) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs69761.2026.11613236
LLM-Assisted Policy Adaptation for Proactive Companion Robots with Continuous Observation
Su, Z., Anciso-Chavez, M., & Grant, G. (2026) · International Conference on Mechatronics, Robotics, and Artificial Intelligence (MRAI) · doi:10.1109/mrai70020.2026.11621744
“In real-world deployments, lightweight offline models such as YOLO [19] and Moondream2 [20] can be executed locally to detect environmental cues.”
Anomaly Detection and Recognition in Complex Commercial Kitchen Environments
Yang, S., Wang, Q., Zhou, Y., & Lai, H. (2026) · IEEE Access · doi:10.1109/access.2026.3717751
Susanto, D. D., Harjanto, T. D., Oei, V., & Jabar, B. A. (2026) · IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology (IAICT) · doi:10.1109/iaict71158.2026.11620669
Research on Layered State Recognition Method of Mixed-Bed Resin based on Improved Mask R-CNN
Shang, J., Xiong, P., Huang, W., et al. (2026) · International Conference on Computer and Communication Systems (ICCCS) · doi:10.1109/icccs69761.2026.11613288
Zeng, B., Li, D., Zhang, W., Wang, J., Liu, H., Zhang, T., Chen, L., Shi, Y., & Jiang, H. (2026) · International Conference on Mechatronics, Robotics, and Artificial Intelligence (MRAI) · doi:10.1109/mrai70020.2026.11621613
“This "foreground-background feature entanglement" phenomenon makes it difficult for traditional single-stage detectors (such as the YOLO series [1,3] ) to construct effective discriminative boundaries with their deep convolutional features, causing classifiers to generate a large number of false…”
InfluentialLightweight Deep Learning Architecture For Edge-Based Road Damage Monitoring
Niharika, D., Akhil, T., & Bhargavi, M. (2026) · 2026 6th International Conference on Intelligent Technologies (CONIT) · doi:10.1109/conit69683.2026.11620831
QS-YOLOv8: a lightweight and efficient detector for small objects
Xie, S., Li, H., Ye, W., Chen, J., Xiong, A., & Zhang, P. (2026) · The Visual Computer · doi:10.1007/s00371-026-04659-9
An AI Based Real Time Automatic Coffee Berry Classification
Ramesh Kumar, P., Lakshmanan, G., Rajesh, V., et al. (2026) · International Conference on Intelligent Technologies (CONIT) · doi:10.1109/conit69683.2026.11620635
“YOLO is a deep learning–based al-gorithm that performs detection in a single pass through a convolutional neural network, enabling fast and efficient real-time performance [7].”
Real-Time Object Detection with YOLOv8: A Deep Learning Approach for Everyday Object Recognition
Okomo, R. B. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621307
“…algorithms to improve temporal consistency while preserving detector efficiency, a paradigm often required in complex real-world scenes J.Redmon et al.[4] demonstrate that targeted fine-tuning of the detection head can yield task-specific accuracy improvements while maintaining real-time operation.”
가상 프로덕션에서의 시각 언어 모델을 이용한 각본 기반 카메라 촬영 자동화
Han, J., & Kim, Y. J. (2026) · Journal of the Korea Computer Graphics Society · doi:10.15701/kcgs.2026.32.3.121
Multi-source data-driven machine learning for lung cancer: diagnosis, treatment, and prognosis
Zou, Y., Zhu, J., Cheng, J., Yan, J., Dai, B., Han, J., & Guo, J. (2026) · Intelligent Opto-Electronics · doi:10.67704/ioe.2026.260006
Yu, C., Li, L., & Huang, B. (2026) · AgriEngineering · doi:10.3390/agriengineering8080311
WCB-YOLO: An Improved Detection Algorithm for Underwater Fishing Vessel Surface Defects
Wei, S., Cao, R., & Sui, J. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621181
“One-stage detectors, represented by YOLO [5] and SSD [6], offer real-time inference speed but often at the cost of reduced accuracy.”
Egsmnet: a lightweight framework for personnel and vehicle detection in UAV emergency rescue imagery
Xiao, P., Wen, Z., Wu, M., & Li, W. (2026) · Pattern Analysis and Applications · doi:10.1007/s10044-026-01746-4
“Typical single-stage representatives include the YOLO [ 4 ] and SSD [ 5], whereas the two-stage camp is exemplified by the R-CNN [ 6] lineage, notably Faster R-CNN [ 7 ] and Cascade R-CNN [ 8].”
URD-YOLO: An Improved YOLO Framework for Underwater Object Detection
Zhang, R., & Sun, M. (2026) · International Conference on Artificial Intelligence, Big Data and Algorithms (CAIBDA) · doi:10.1109/caibda70336.2026.11621102
“Among existing methods, the YOLO series (You Only Look Once) [1] is a representative one-stage detector.”
Camera-based Object Detection with GPS Tagging
Fatima, S., Subramanyam, M. V., Gali, S., et al. (2026) · International Conference Computing Methodologies and Communication · doi:10.1109/iccmc69250.2026.11624718
“Redmon et al. [1] proposed the YOLO (You Only Look Once) architecture, which unified object localization and classification into a single-stage detection framework, enabling real-time object detection with high processing speed.”
Mukkapati, R. K., Nelakurthi, J., Prathap Kumar, R., et al. (2026) · International Conference on Computing Methodologies and Communication (ICCMC) · doi:10.1109/iccmc69250.2026.11624963
“Segmentation and object detection models such as U-Net [9] and YOLO [10], [11] can further support disease localization or fast prediction.”
Detection Methods and Condition Assessment of High-Voltage Bushings: A Systematic Review
Wu, G., Wei, S.-J., Shi, Z., et al. (2026) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2026.3718089
A Progressive Frequency-Guided Decomposition Method for Infrared and Visible Image Fusion
Sun, L., Zhang, M., & Fang, E. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3714861
“We adopt the YOLOv5s detector [47] on the annotated MSRS subset to evaluate the contribution of fused images to object detection.”
Advances in Deep Learning and Computer Vision for Sewer Inspection: A State-of-the-Art Review
Tiwari, N., Shammout, D. Z., & Jalali, H. H. (2026) · Journal of Pipeline Systems Engineering and Practice · doi:10.1061/jpsea2.pseng-2258
Thant, P. T., Endo, T., Mori, Y., Hirai, N., Takei, M., Yumine, T., Munetomo, M., & Matsuo, Y. (2026) · IEEE International Symposium on Cluster, Cloud and Internet Computing Workshops (CCGridW) · doi:10.1109/ccgridw69005.2026.00043
“During our test cases, Otsu’s thresholding which described in [15] was applied for morphology detection, YOLO [11], specifically YOLOv8, was used for particle detection, and core Python functions were employed for the measurement tool.”
Hoang, T.-H., Hoang, Q.-V., Pham, N.-T., et al. (2026) · Journal of Visual Communication and Image Representation · doi:10.1016/j.jvcir.2026.104910
Object detection in histology: A multi-dataset benchmark and test-time inference
Leordean, D.-V., & Ardelean, E.-R. (2026) · PLoS ONE · doi:10.1371/journal.pone.0354618
SCD-YOLO: A Detail-Enhanced Network for Strip Steel Surface Defect Detection
Jiang, J., & Su, Y. (2026) · Open Journal of Applied Sciences · doi:10.4236/ojapps.2026.167145
EAA: Adapting the Segment Anything Model for Perception-driven Astronomical Source Extraction
Wang, H., Zhang, T., Chen, S., Zhang, G., Shen, Z., Yang, X., Wheatley, P. J., Fernández, J. F., Apergis, I., Burleigh, M. R., et al. (2026) · The Astrophysical Journal Supplement Series · doi:10.3847/1538-4365/ae7b24 · Cited by 1
BladeYOLO: Wind Turbine Blade Defect Detection With Limited Annotations and Weak-Saliency Awareness
Xu, Y., Zhang, F., Wang, F., et al. (2026) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2026.3717985
“SSD [27] introduced multiscale prediction strategies, while YOLO [28] formulated object detection as a single regression problem and became a representative real-time detector.”
LocSAM: Modular SAM enhancement for dense object localization in complex scenes
Wang, J., Zhang, Y., Li, B., Hu, Y., Zhang, B., & Yin, B. (2026) · Computer Vision and Image Understanding · doi:10.1016/j.cviu.2026.104878
Developing a Low Resource Intensive Algorithm for Early Detection of Flying Swarm of Pest BirdsPreprint
Mosoeu, J. R., Serutla, L., & Mwanza, A. (2026) · Research Square · doi:10.21203/rs.3.rs-10360131/v1
IoT-based Smart Helmet for Firefighters with Machine Learning-based Human, Fire and Smoke Detection
Sri Silpa, P., Anusha, S. L., Haritha, K., Harshita, M., & Sai, S. M. (2026) · International Conference on Computing Methodologies and Communication (ICCMC) · doi:10.1109/iccmc69250.2026.11624823
“The YOLO series [6], which began with the pioneering work of Redmon et al., has seen great progress through subsequent versions.”
CMMS-DETR: Cross-Modal Multiscale Feature Fusion Network for UAV Remote Sensing Detection
Wang, Y., Mu, J., Yang, Q., Gao, G., He, J., Yu, J., & Chou, X. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3718002 · Cited by 1
TunVECM: vector graph structured extraction and intelligent checking for tunnel portal
Li, Y., Wei, S., Chang, Y., Qi, C., & Li, Y. (2026) · Tunnelling and Underground Space Technology · doi:10.1016/j.tust.2026.107966
Scalable Object Detection in the Car Interior With Vision Foundation Models
Schmidt, S., Mészáros, B., Firintepe, A., & Günnemann, S. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11623817
Autonomous Driving on Forest Roads under Unreliable GNSS *
Didari, H., Wipfler, H.-P., Regner, S., Oberpertinger, J., Fraundorfer, F., & Steinbauer-Wagner, G. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11623967
“We adopt a YOLOv11-based segmentation model [9], [10] that performs pixel-wise classification of forest-road imagery into road surface, road boundaries, and background in a single forward pass.”
Zhang, Y., Pan, R., Zhang, Y., et al. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.114540
Qin, D., Gan, S., Xu, K., et al. (2026) · IET Intelligent Transport Systems · doi:10.1049/itr2.70297
Localized SHAP Explainability for Object Detection Based on Parallel Attribution Analysis
Trofimov, Y. V., Averkin, A. N., Shevchenko, A. V., Трусов, И. А., Алексеев, А. К., & Kondrashova, E. S. (2026) · IEEE International Conference on Soft Computing and Measurements (SCM) · doi:10.1109/scm71573.2026.11625907
“Deep neural networks demonstrate high accuracy in image-based object detection tasks [1, 2]; however, their practical deployment in sensitive domains—medical diagnostics, industrial quality control, and autonomous transportation—is hindered by the lack of transparency in the decisions made.”
Optimized Small Object Detection Model for Autonomous Driving
Ahn, G., Bong, E.-J., & Kee, S.-C. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11623867
“YOLOv3 [3] introduced multi-scale detection using three feature maps and the Darknet-53 backbone, improving performance over previous YOLO versions.”
Hakimi, O., Liu, H., & Abudayyeh, O. Y. (2026) · Journal of Computing in Civil Engineering · doi:10.1061/jccee5.cpeng-7695
Lane Marking Segmentation Based on Conditional Diffusion Model
Liu, J., Li, J. Q., Du, X., & Golalipour, A. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11624100 · Cited by 1
“…lane marking segmentation is formulated as a denoising diffusion mapping from noisy masks to lane marking masks. architectures, such as YOLO [7] and U-Net [8] variants, (ii) Transformer architectures [9], and (iii) generative approaches, including Generative Adversarial Network (GAN) [10]…”
Vinh, T. Q., & Pham, H. L. (2026) · Multimedia Tools and Applications · doi:10.1007/s11042-026-21823-4
“The presentation for this type of algorithm is the YOLO (You Only Look Once) series [9] and SSD (Single Shot Detectors) [10].”
Voice-Controlled Robotic Arm System for Tabletop Manipulation via Large Language Model and 3D Vision
Thavornvong, S., Kitsommart, A., Thabua, M., Thanaluck, C., & Yuenyong, S. (2026) · International Conference on Ubiquitous Robots (UR) · doi:10.1109/ur69537.2026.11626685
“: The YOLO family [4], [5] provides real-time detection with segmentation variants enabling per-pixel masks.”
Multi-Opponent Tracking with Dual Stage Association for Autonomous Racing
Dahal, P., & Morris, D. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11624010
“Additionally, we use 2D-to-3D projection techniques with the 2D detector YOLO [17] to further supplement long-range perception.”
Utilizing YOLOv7 Algorithm for Gossypium sp. Detection: An Implementation Study
Anggraini, B. S. K., Utari, D. T., & Hakim, R. B. F. (2026) · Springer eBooks · doi:10.1007/978-3-032-32458-0_10
Cheong, M. Y. Y., Xu, X., Wang, L. R., Yeo, T. X. J., Shanmugam, H., Lim, S. P., Fan, B. E., & Fan, X. (2026) · PLOS Digital Health · doi:10.1371/journal.pdig.0001514
“DREAM utilises the You Only Look Once (YOLO) model [16] for precise blood cell detection and classification, followed by segmentation using K-Means Clustering and the Segment Anything Model (SAM2) [17,18] to isolate cellular components.”
Fahry, M., & Wiranto, A. (2026) · Indonesian Journal of Innovation Multidisipliner Research · doi:10.69693/ijim.v4i3.1834
“( Joseph Redmon, Santosh Divvala, 2016).”
VisionGuard: AI Powered Examination Integrity and Proctoring Management System
Priyanka Rao B, M. (2026) · International Journal of Advanced Research in Science Communication and Technology · doi:10.48175/ijarsct-37773
Bounded Deterministic Replay for Asynchronous Acoustic–Visual UAS Tracking
Tsai, S.-E. (2026) · IEEE Access · doi:10.1109/access.2026.3718076
Efficient Cross-Country Data Acquisition Strategy for ADAS via Street-View Imagery
Wu, Y., Slieter, D., Esselborn, C., Abouelazm, A., Tseng, T. Y., & Zöllner, J. M. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11624023
“TSD can be formulated as a standard object detection problem, where models such as YOLO [9], Faster R-CNN [10], and RetinaNet [11] are commonly preferred over transformer-based models [12] due to their favorable accuracy-speed trade-offs.”
Yang, C., Cao, C., Yan, J., Wang, G., Zhang, X., Ding, Y., Ma, C., & Liu, Y. (2026) · Journal of Forestry Research · doi:10.1007/s11676-026-02114-y
“One-stage detectors, especially those in the You Only Look Once (YOLO) family (Redmon et al. 2016), have consequently become the mainstream choice.”
Drivers’ Visual Attention During Repeated L2-Driving: Analyzing Pedestrian Crossing Scenarios
Nkusi, E. M., Aragón, C. M., Marberger, C., & Bengler, K. J. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11623955
“Architectures such as the R-CNN family [24] and You Only Look Once (YOLO) [25] enable high detection accuracy and real-time inference.”
Akylbekov, O., Zakariya, G., Bekarystankyzy, A., Kassenkhan, A. M., Amantay, B., & Жарылкасынович, Б. Т. (2026) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981261461488
Yurdakul, M., Zeynalov, J., Ersoy, M. T., Özger, F., & Paçal, İ. (2026) · Diagnostics · doi:10.3390/diagnostics16152407
“The original YOLO framework was developed for real-time object detection [8].”
InfluentialConcrete Crack Segmentation Algorithm Based on Hybrid-Attention Feature Enhancement
Yan, T., Zhang, X., Li, P., et al. (2026) · Buildings · doi:10.3390/buildings16153031
LDFpose: A lightweight deep fusion network for 6D occlusion object pose estimation
Dong, G., Wu, Y., Dai, L., Chang, H., Tang, Z., & Liu, D. (2026) · Image and Vision Computing · doi:10.1016/j.imavis.2026.106147
Early Diagnosis of Autism through Behaviors from Semi-Structured Triadic Play Interaction Videos
Berlin, S. J., Choudhury, A. R., Prasad, A. N., Murli, S., Jayagopi, D. B., Raman, V., & Rajagopalan, S. S. (2025) · Indian Conference on Computer Vision, Graphics and Image Processing · doi:10.1145/3774521.3774543 · Cited by 1
“• Person and Object Detection : As an initial step, persons and objects are detected using YOLOv7 [59].”
Benchmarking Object Detection Models for Tuta Absoluta Egg Identification in Field Images
Katranas, G., Kavasidis, I., Tsiropoulos, Z., et al. (2026) · doi:10.1109/smartcomp-companion70724.2026.00048
“In contrast, one-stage detectors, including YOLO [10], RetinaNet [2], and FCOS [3], perform object localization and classification in a single forward pass, offering improved efficiency while maintaining competitive accuracy.”
Li, Z., Huang, M., Zhu, H., Tang, Y., Zhao, J., & Zhang, Q. (2026) · International Journal of Rock Mechanics and Mining Sciences · doi:10.1016/j.ijrmms.2026.106655
A U-Net architecture with graph attention networks to accurately define tooth boundaries
Akefi, E., & Khotanlou, H. (2025) · Conference on Information and Knowledge Technology · doi:10.1109/ikt69652.2025.11626688
“…research demonstrates that algorithms utilizing deep learning, including U-Net and its variants such as Attention U-Net and U-Net++ [2], alongside object detection models like Mask R-CNN [3] and YOLO [4], have been primarily employed for the segmentation and identification of teeth in radiographs.”
Automated Litter Characterization for Smart Cities
Diaz-Silva, E., Gao, G., Begur, N., Charles, S., Arra, R., Chatterjee, S., Zarif, M., & Hallstrom, J. O. (2026) · IEEE International Conference on Smart Computing Workshops and Other Affiliated Events (SMARTCOMP Companion) · doi:10.1109/smartcomp-companion70724.2026.00059
“By contrast, single-frame CNNs, including YOLO [15], SSD [16], and FR-CNN [17] are particularly well-suited to real-time applications.”
da Costa, E. R., Lemos, C., Cáceres, E. A. M. D. R., et al. (2026) · doi:10.5753/wcama.2026.19543
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.