Full search page →

YOLO v1: detection as a single regression

Optional · paper guide 1 of 22 for DLCV 03A · One-Stage Detection, Read Through YOLOv4. Citation numbers refer to the reference list at the bottom of this page.

Navigate: this page and this track: Navigate: this page and this track

On this page

Supplementary notes · 2

  1. Ultralytics releases without a paper
  2. Folding a re-parameterized block into one convolution

Paper guides · 22

  1. YOLO v1: detection as a single regression
  2. YOLOv2 and YOLO9000: anchors, priors and joint training
  3. YOLOv3: three scales and Darknet-53
  4. SSD: multi-scale default boxes
  5. RetinaNet and the focal loss
  6. FCOS: per-pixel anchor-free detection
  7. CenterNet: objects as points
  8. YOLOR: explicit features and learned implicit representations
  9. YOLOX: anchor-free YOLO with SimOTA
  10. YOLOv6: a re-parameterized industrial detector
  11. YOLOv7: a trainable bag of freebies
  12. YOLOv9: programmable gradient information
  13. YOLOv10: NMS-free training with dual assignments
  14. YOLOv12: an attention-centric YOLO
  15. YOLOv13: hypergraph correlation
  16. YOLO26: a DFL-free end-to-end YOLO
  17. RT-DETR: a real-time DETR
  18. RT-DETRv2: a bag of freebies for RT-DETR
  19. D-FINE: boxes as refined distributions
  20. DEIM: dense one-to-one matching
  21. RF-DETR: architecture search on a DINOv2 backbone
  22. DEIMv2: DINOv3 features for real-time DETRs

Deep Learning for Computer Vision

  1. 02 · Deep Image Classification
View the whole course tree →

Paper: J. Redmon, S. Divvala, R. Girshick and A. Farhadi, “You only look once: Unified, real-time object detection,” CVPR 2016 [7]. Related part of YOLOv4: the detector anatomy.

Challenge

In 2015 the accurate detectors were pipelines: a proposal method suggested about 2,000 candidate regions, a convolutional network classified each, and further stages refined boxes and removed duplicates. Each stage had to be tuned separately, and the systems were slow; the paper notes that R-CNN took more than 40 seconds per image at test time, and that among the GPU detectors of the time only a fast deformable-parts implementation ran at 30 FPS or more [7]. Fast R-CNN [69] and Faster R-CNN [70] made the pipeline faster but not real-time. The challenge was a detector that is real-time by design, trained end to end on the detection objective.

Method

Follow one image through convolutional features and the final grid prediction; this original architecture is not interchangeable with later multi-scale YOLO heads.
Original paper figure 3 — Follow one image through convolutional features and the final grid prediction; this original architecture is not interchangeable with later multi-scale YOLO heads.

Figure source: [7], arXiv:1506.02640v5, PDF page 3. Original PDF.

One grid, one tensor. The image is resized to 448×448448 \times 448 and divided into an S×SS \times S grid. The cell that contains an object’s centre is responsible for that object. Each cell predicts BB boxes, each with five numbers (x,y,w,h,confidence)(x, y, w, h, \text{confidence}), and one set of CC conditional class probabilities Pr⁡(Classi∣Object)\Pr(\text{Class}_i \mid \text{Object}), so the whole output is a tensor of shape

S×S×(5B+C),S \times S \times (5B + C),

which for PASCAL VOC [71] (S=7S = 7, B=2B = 2, C=20C = 20) is 7×7×307 \times 7 \times 30 [7]. The box centre (x,y)(x, y) is an offset within the cell, between 0 and 1; ww and hh are relative to the whole image. The confidence is trained toward Pr⁡(Object)⋅IoUpredtruth\Pr(\text{Object}) \cdot \text{IoU}^{\text{truth}}_{\text{pred}}, zero when no object is present. At test time the class-specific score of each box is

Pr⁡(Classi∣Object)⋅Pr⁡(Object)⋅IoUpredtruth=Pr⁡(Classi)⋅IoUpredtruth,\Pr(\text{Class}_i \mid \text{Object}) \cdot \Pr(\text{Object}) \cdot \text{IoU}^{\text{truth}}_{\text{pred}} = \Pr(\text{Class}_i) \cdot \text{IoU}^{\text{truth}}_{\text{pred}},

which encodes both the class probability and how well the box fits [7]. The network therefore makes only 7×7×2=987 \times 7 \times 2 = 98 box predictions per image, against about 2,000 proposals for a selective-search pipeline.

Network. 24 convolutional layers (inspired by GoogLeNet, with 1×11 \times 1 reduction layers followed by 3×33 \times 3 convolutions) and 2 fully connected layers; “Fast YOLO” uses 9 convolutional layers with fewer filters [7]. The first 20 convolutional layers are pre-trained on ImageNet [30] at 224×224224 \times 224 (about a week of training, 88% single-crop top-5 accuracy); four convolutional and two fully connected layers are added for detection and the input resolution is doubled to 448×448448 \times 448. All layers except the last use the leaky rectifier ϕ(x)=x\phi(x) = x for x>0x \gt 0 and 0.1x0.1x otherwise; the last layer is linear [7].

Loss. The paper optimizes a sum of squared errors with two corrections. Most cells contain no object, so their confidence errors would overwhelm the others: the coordinate terms are weighted by λcoord=5\lambda_{\text{coord}} = 5 and the no-object confidence terms by λnoobj=0.5\lambda_{\text{noobj}} = 0.5. Errors in large boxes should matter less than in small ones, so the network predicts w\sqrt{w} and h\sqrt{h}. The loss is

L=λcoord∑i=0S2∑j=0B1ijobj[(xi−x^i)2+(yi−y^i)2+(wi−w^i)2+(hi−h^i)2]+∑i=0S2∑j=0B1ijobj(Ci−C^i)2+λnoobj∑i=0S2∑j=0B1ijnoobj(Ci−C^i)2+∑i=0S21iobj∑c(pi(c)−p^i(c))2,\begin{aligned} \mathcal{L} ={}& \lambda_{\text{coord}} \sum_{i=0}^{S^2} \sum_{j=0}^{B} \mathbb{1}^{\text{obj}}_{ij} \Big[ (x_i - \hat{x}_i)^2 + (y_i - \hat{y}_i)^2 + \big(\sqrt{w_i} - \sqrt{\hat{w}_i}\big)^2 + \big(\sqrt{h_i} - \sqrt{\hat{h}_i}\big)^2 \Big] \\ &+ \sum_{i=0}^{S^2} \sum_{j=0}^{B} \mathbb{1}^{\text{obj}}_{ij} (C_i - \hat{C}_i)^2 + \lambda_{\text{noobj}} \sum_{i=0}^{S^2} \sum_{j=0}^{B} \mathbb{1}^{\text{noobj}}_{ij} (C_i - \hat{C}_i)^2 + \sum_{i=0}^{S^2} \mathbb{1}^{\text{obj}}_{i} \sum_{c} \big(p_i(c) - \hat{p}_i(c)\big)^2, \end{aligned}

where ii indexes cells and jj the boxes of a cell, 1iobj\mathbb{1}^{\text{obj}}_{i} is 1 if an object appears in cell ii, 1ijobj\mathbb{1}^{\text{obj}}_{ij} is 1 if box predictor jj of cell ii is “responsible” for that object (the predictor with the highest current IoU with it), 1ijnoobj\mathbb{1}^{\text{noobj}}_{ij} is its complement, hatted symbols are predictions and unhatted ones targets, CC is the confidence and pi(c)p_i(c) the class probability [7]. Because only the responsible predictor is trained on each object, the two predictors of a cell specialize in different sizes and shapes; this is the first YOLO label-assignment rule.

Training. About 135 epochs on VOC 2007 and 2012 train and validation data, batch 64, momentum 0.9, weight decay 0.0005; the learning rate is raised from 10−310^{-3} to 10−210^{-2} over the first epochs, kept for 75 epochs, then 10−310^{-3} for 30 and 10−410^{-4} for 30. Dropout [48] with rate 0.5 after the first fully connected layer, random scaling and translations of up to 20% of the image size, and random exposure and saturation changes of up to a factor 1.5 in HSV space [7]. Greedy NMS is still applied at test time; the paper reports that it adds 2–3% mAP.

Results

On PASCAL VOC 2007 test (trained on 2007+2012), YOLO reaches 63.4% mAP at 45 FPS and Fast YOLO 52.7% mAP at 155 FPS, on a Titan X GPU without batching; the real-time deformable-parts baselines reach 26.1% at 30 FPS and 16.0% at 100 FPS [7]. Slower detectors are more accurate: Fast R-CNN 70.0% at 0.5 FPS, Faster R-CNN with VGG-16 73.2% at 7 FPS, and a YOLO trained with a VGG-16 backbone 66.4% at 21 FPS [7]. On VOC 2012 test YOLO scores 57.9% mAP [7]. The headline is speed: more than twice the mAP of earlier real-time systems.

Ablations

The paper has no component-by-component ablation table; its analyses play that role. An error breakdown on VOC 2007 (following a standard diagnosis tool) shows that 19.0% of YOLO’s top detections are localization errors and 4.75% background errors, against 8.6% and 13.6% for Fast R-CNN: YOLO localizes worse but mistakes background for objects about three times less often [7]. Using YOLO to rescore Fast R-CNN’s detections raises Fast R-CNN from 71.8% to 75.0% mAP, which the paper attributes to these different error types rather than to simple ensembling [7]. A person-detection test on artwork shows that YOLO’s accuracy drops less than that of R-CNN-style detectors when moving from photographs to paintings [7].

Conclusion

YOLO v1 established detection as a single regression from image to boxes and classes, trained end to end, and showed that such a detector can run at video rate with useful accuracy. Every later YOLO keeps its “look once, predict densely on a grid” structure.

Problems

The paper lists its own limits [7]: each cell predicts only two boxes and one class, so nearby small objects (a flock of birds) compete for one cell; boxes are learned from data with no shape prior, so unusual aspect ratios generalize poorly; coarse features after many downsampling steps limit localization; and the squared-error loss treats small and large boxes alike, despite the square-root correction. Localization was its main source of error. YOLOv2 answered with anchors, batch normalization and a fully convolutional design; SSD answered with predictions on several feature maps.

References

  1. C.-Y. Wang, A. Bochkovskiy and H.-Y. M. Liao, “YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors,” in Proc. IEEE/CVF CVPR, pp. 7464–7475, 2023. doi
  2. C.-Y. Wang, I-H. Yeh and H.-Y. M. Liao, “YOLOv9: Learning what you want to learn using programmable gradient information,” in Proc. ECCV, 2024. doi
  3. G. Jocher, “Ultralytics YOLOv5,” software (first released 2020; v7.0 released 22 November 2022). release · documentation
  4. G. Jocher, A. Chaurasia and J. Qiu, “Ultralytics YOLOv8,” software, version 8.0.0, 2023 (no paper). docs · code
  5. G. Jocher and J. Qiu, “Ultralytics YOLO11,” software, version 11.0.0, 2024 (no paper). docs · code
  6. J. Lu, J. Jia, J. Yawl and C. Zhang, “YOLOv14: Adaptive real-time object detection for diverse imaging conditions,” arXiv:2608.04720, 2026 (withdrawn by the authors, 20 August 2026). arXiv
  7. J. Redmon, S. Divvala, R. Girshick and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proc. IEEE CVPR, pp. 779–788, 2016. doi · arXiv
  8. J. Redmon and A. Farhadi, “YOLO9000: Better, faster, stronger,” in Proc. IEEE CVPR, pp. 6517–6525, 2017. doi · arXiv
  9. J. Redmon and A. Farhadi, “YOLOv3: An incremental improvement,” tech. report, arXiv:1804.02767, 2018. arXiv
  10. W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. Reed, C. Fu, A. C. Berg, “SSD: Single Shot MultiBox Detector,” Lecture Notes in Computer Science / Computer Vision – ECCV 2016, 2016. doi · manuscript
  11. T.-Y. Lin, P. Goyal, R. Girshick, K. He and P. Dollár, “Focal loss for dense object detection,” in Proc. IEEE ICCV, pp. 2999–3007, 2017. doi
  12. Z. Tian, C. Shen, H. Chen, T. He, “FCOS: Fully Convolutional One-Stage Object Detection,” 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019. doi · manuscript
  13. X. Zhou, D. Wang, P. Krähenbühl, “Objects as Points,” arXiv preprint, 2019. source
  14. C. Wang, I. Yeh, H. M. Liao, “You Only Learn One Representation: Unified Network for Multiple Tasks,” arXiv preprint, 2021. source
  15. Z. Ge, S. Liu, F. Wang, Z. Li and J. Sun, “YOLOX: Exceeding YOLO series in 2021,” arXiv:2107.08430, 2021. arXiv
  16. C. Li, L. Li, H. Jiang, K. Weng, Y. Geng et al., “YOLOv6: A single-stage object detection framework for industrial applications,” tech. report, arXiv:2209.02976, 2022. arXiv
  17. A. Wang, H. Chen, L. Liu, K. Chen, Z. Lin, J. Han and G. Ding, “YOLOv10: Real-time end-to-end object detection,” in Proc. NeurIPS, 2024. arXiv
  18. Y. Tian, Q. Ye and D. Doermann, “YOLOv12: Attention-centric real-time object detectors,” in Advances in Neural Information Processing Systems 38 (NeurIPS), 2025. proceedings · arXiv
  19. M. Lei, S. Li, Y. Wu, H. Hu, Y. Zhou et al., “YOLOv13: Real-time object detection with hypergraph-enhanced adaptive visual perception,” arXiv:2506.17733, 2025. arXiv
  20. G. Jocher, J. Qiu, M. Liu, S. Lyu, F. C. Akyon and M. E. Kalfaoglu, “Ultralytics YOLO26: Unified real-time end-to-end vision models,” arXiv:2606.03748, 2026 (software released January 2026). arXiv · docs
  21. Y. Zhao, W. Lv, S. Xu, J. Wei, G. Wang, Q. Dang, Y. Liu, J. Chen, “DETRs Beat YOLOs on Real-time Object Detection,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024. source
  22. W. Lv, Y. Zhao, Q. Chang, K. Huang, G. Wang and Y. Liu, “RT-DETRv2: Improved baseline with bag-of-freebies for real-time detection transformer,” arXiv:2407.17140, 2024. arXiv
  23. Y. Peng, H. Li, P. Wu, Y. Zhang, X. Sun and F. Wu, “D-FINE: Redefine regression task in DETRs as fine-grained distribution refinement,” in Proc. ICLR, 2025. arXiv · code
  24. S. Huang, Z. Lu, X. Cun, Y. Yu, X. Zhou, X. Shen, “DEIM: DETR with Improved Matching for Fast Convergence,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025. source
  25. I. Robinson, P. Robicheaux, M. Popov, D. Ramanan and N. Peri, “RF-DETR: Neural architecture search for real-time detection transformers,” in Proc. ICLR, 2026. arXiv
  26. S. Huang, Y. Hou, L. Liu, X. Yu and X. Shen, “Real-time object detection meets DINOv3,” arXiv:2509.20787, 2025 (revised 2026). arXiv
  27. A. Bochkovskiy, C.-Y. Wang and H.-Y. M. Liao, “YOLOv4: Optimal speed and accuracy of object detection,” arXiv:2004.10934, 2020. arXiv
  28. M. Tan, R. Pang, Q. V. Le, “EfficientDet: Scalable and Efficient Object Detection,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020. doi · manuscript
  29. NVIDIA, “TensorRT documentation,” official documentation. link
  30. J. Deng, W. Dong, R. Socher, L. Li, Kai Li, Li Fei-Fei, “ImageNet: A large-scale hierarchical image database,” 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009. source
  31. K. Simonyan, A. Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” arXiv preprint, 2014. source
  32. K. He, X. Zhang, S. Ren, J. Sun, “Deep Residual Learning for Image Recognition,” arXiv preprint, 2015. source
  33. T.-Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan and S. Belongie, “Feature pyramid networks for object detection,” in Proc. IEEE CVPR, 2017. doi
  34. S. Liu, L. Qi, H. Qin, J. Shi and J. Jia, “Path aggregation network for instance segmentation,” in Proc. IEEE/CVF CVPR, pp. 8759–8768, 2018. doi
  35. K. He, X. Zhang, S. Ren, J. Sun, “Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition,” Lecture Notes in Computer Science / Computer Vision – ECCV 2014, 2014. doi · manuscript
  36. C.-Y. Wang, H.-Y. M. Liao, I-H. Yeh, Y.-H. Wu, P.-Y. Chen and J.-W. Hsieh, “CSPNet: A new backbone that can enhance learning capability of CNN,” CVPR Workshops, 2020. arXiv · doi
  37. A. Bochkovskiy and contributors, “Darknet: YOLOv4 configuration and detection layer,” official implementation, accessed 2 November 2026. configuration · assignment and decoding
  38. M. Tan, Q. V. Le, “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” ICML, 2019. source · proceedings
  39. G. Huang, Z. Liu, L. Van Der Maaten, K. Q. Weinberger, “Densely Connected Convolutional Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. doi · manuscript
  40. S. Xie, R. Girshick, P. Dollar, Z. Tu, K. He, “Aggregated Residual Transformations for Deep Neural Networks,” 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017. doi · manuscript
  41. D. Misra, “Mish: A Self Regularized Non-Monotonic Activation Function,” BMVC, 2020. source
  42. P. Ramachandran, B. Zoph, Q. V. Le, “Searching for Activation Functions,” arXiv preprint, 2017. source
  43. J. Yu, Y. Jiang, Z. Wang, Z. Cao, T. Huang, “UnitBox: An Advanced Object Detection Network,” ACM Multimedia, 2016. source
  44. H. Rezatofighi, N. Tsoi, J. Gwak, A. Sadeghian, I. Reid, S. Savarese, “Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression,” arXiv preprint, 2019. source
  45. Z. Zheng, P. Wang, W. Liu, J. Li, R. Ye and D. Ren, “Distance-IoU loss: Faster and better learning for bounding box regression,” in Proc. AAAI, pp. 12993–13000, 2020. doi
  46. H. Zhang, M. Cisse, Y. N. Dauphin and D. Lopez-Paz, “mixup: Beyond empirical risk minimization,” in Proc. ICLR, 2018. arXiv
  47. S. Yun, D. Han, S. Chun, S. J. Oh, Y. Yoo, J. Choe, “CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features,” 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019. doi · manuscript
  48. Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov, “Dropout: A Simple Way to Prevent Neural Networks from Overfitting,” Journal of Machine Learning Research, 2014. source
  49. G. Ghiasi, T. Lin, Q. V. Le, “DropBlock: A regularization method for convolutional networks,” NeurIPS, 2018. source · proceedings
  50. C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, Z. Wojna, “Rethinking the Inception Architecture for Computer Vision,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. doi · manuscript
  51. S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift,” in Proc. ICML, PMLR vol. 37, pp. 448–456, 2015. link
  52. Z. Yao, Y. Cao, S. Zheng, G. Huang, S. Lin, “Cross-Iteration Batch Normalization,” 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021. doi · manuscript
  53. S. Woo, J. Park, J. Lee, I. S. Kweon, “CBAM: Convolutional Block Attention Module,” Lecture Notes in Computer Science / Computer Vision – ECCV 2018, 2018. doi · manuscript
  54. I. Loshchilov, F. Hutter, “SGDR: Stochastic Gradient Descent with Warm Restarts,” ICLR, 2017. source
  55. T.-Y. Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollár and C. L. Zitnick, “Microsoft COCO: Common objects in context,” in Proc. ECCV, pp. 740–755, 2014. doi
  56. COCO Consortium, “COCO detection evaluation,” official evaluation description. link
  57. ONNX, “Open Neural Network Exchange,” official project page. link
  58. B. Jacob, S. Kligys, B. Chen, M. Zhu, M. Tang, A. Howard, H. Adam and D. Kalenichenko, “Quantization and training of neural networks for efficient integer-arithmetic-only inference,” in Proc. IEEE/CVF CVPR, pp. 2704–2713, 2018. doi
  59. N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov and S. Zagoruyko, “End-to-end object detection with transformers,” in Proc. ECCV, 2020. doi
  60. H. W. Kuhn, “The Hungarian method for the assignment problem,” Naval Research Logistics Quarterly, vol. 2, pp. 83–97, 1955. doi
  61. J. Terven, D.-M. Córdova-Esparza and J.-A. Romero-González, “A comprehensive review of YOLO architectures in computer vision: From YOLOv1 to YOLOv8 and YOLO-NAS,” Machine Learning and Knowledge Extraction, vol. 5, no. 4, pp. 1680–1716, 2023. doi · arXiv
  62. N. Jegham, C. Y. Koh, M. Abdelatti and A. Hendawi, “YOLO evolution: A comprehensive benchmark and architectural review of YOLOv12, YOLO11, and their previous versions,” arXiv:2411.00201, 2024 (revised 2025). arXiv
  63. R. Sapkota and M. Karkee, “Ultralytics YOLO evolution: An overview of YOLO27, YOLO26, YOLO11, YOLOv8, and YOLOv5 object detectors for computer vision and pattern recognition,” arXiv:2510.09653, 2025 (revised 2026). arXiv
  64. T. Shehzadi, K. A. Hashmi, M. Liwicki, D. Stricker and M. Z. Afzal, “Object detection with transformers: A review,” Sensors, vol. 25, no. 19, 2025. doi
  65. Ultralytics, “YOLO27 (coming soon),” official documentation page, accessed 2 November 2026. docs
  66. M. Oquab, T. Darcet, T. Moutakanni, H. Vo, M. Szafraniec et al., “DINOv2: Learning robust visual features without supervision,” arXiv:2304.07193, 2023. arXiv
  67. O. Siméoni, H. V. Vo, M. Seitzer, F. Baldassarre, M. Oquab et al., “DINOv3,” arXiv:2508.10104, 2025. arXiv
  68. X. Ding, X. Zhang, N. Ma, J. Han, G. Ding and J. Sun, “RepVGG: Making VGG-style ConvNets great again,” in Proc. IEEE/CVF CVPR, pp. 13728–13737, 2021. doi
  69. R. Girshick, “Fast R-CNN,” 2015 IEEE International Conference on Computer Vision (ICCV), 2015. doi · manuscript
  70. S. Ren, K. He, R. Girshick, J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” arXiv preprint, 2015. source
  71. M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, A. Zisserman, “The Pascal Visual Object Classes (VOC) Challenge,” International Journal of Computer Vision, 2009. source
  72. A. Shrivastava, A. Gupta, R. Girshick, “Training Region-Based Object Detectors with Online Hard Example Mining,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016. doi · manuscript
  73. H. Zhang, Y. Wang, F. Dayoub and N. Sünderhauf, “VarifocalNet: An IoU-aware dense object detector,” in Proc. IEEE/CVF CVPR, pp. 8510–8519, 2021. doi
  74. YOLOX contributors, “YOLOX detection head,” official implementation, accessed 2 November 2026. code
  75. K. Duan, S. Bai, L. Xie, H. Qi, Q. Huang, Q. Tian, “CenterNet: Keypoint Triplets for Object Detection,” arXiv preprint, 2019. source
  76. Z. Ge, S. Liu, Z. Li, O. Yoshie and J. Sun, “OTA: Optimal transport assignment for object detection,” in Proc. IEEE/CVF CVPR, pp. 303–312, 2021. doi
  77. C. Li, L. Li, Y. Geng, et al., “YOLOv6 v3.0: A Full-Scale Reloading,” arXiv preprint, 2023. source
  78. C. Feng, Y. Zhong, Y. Gao, M. R. Scott, W. Huang, “TOOD: Task-Aligned One-Stage Object Detection,” Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021. source
  79. Z. Gevorgyan, “SIoU Loss: More Powerful Learning for Bounding Box Regression,” arXiv preprint, 2022. source
  80. X. Li, W. Wang, L. Wu, S. Chen, X. Hu, J. Li, J. Tang and J. Yang, “Generalized focal loss: Learning qualified and distributed bounding boxes for dense object detection,” in Advances in Neural Information Processing Systems 33 (NeurIPS), 2020. proceedings
  81. X. Ding, H. Chen, X. Zhang, K. Huang, J. Han, G. Ding, “Re-parameterizing Your Optimizers rather than Architectures,” ICLR, 2023. source
  82. S. Zhang, C. Chi, Y. Yao, Z. Lei, S. Z. Li, “Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020. doi · manuscript
  83. C. Wang, H. M. Liao, I. Yeh, “Designing Network Design Strategies Through Gradient Path Analysis,” arXiv preprint, 2022. source
  84. T. Dao, D. Fu, S. Ermon, A. Rudra, C. Ré, “FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness,” Advances in Neural Information Processing Systems 35, 2022. doi · manuscript
  85. D. P. Kingma, J. Ba, “Adam: A Method for Stochastic Optimization,” ICLR, 2015. source
  86. K. Jordan, “Muon: An optimizer for hidden layers in neural networks,” technical blog, 2024. source
  87. S. Shao, Z. Li, T. Zhang, C. Peng, G. Yu, X. Zhang, J. Li, J. Sun, “Objects365: A Large-Scale, High-Quality Dataset for Object Detection,” Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019. source