---
title: "YOLOv4: Optimal Speed and Accuracy of Object Detection (2020)"
description: "The most accurate CNN object detectors are usually too slow for real-time use, and training them… 102 references, cited by 115 papers here."
authors: ["Alexey Bochkovskiy", "Chien-Yao Wang", "Hong-Yuan Mark Liao"]
year: 2020
venue: "arXiv:2004.10934"
source: "https://arxiv.org/abs/2004.10934"
doi: "10.48550/arxiv.2004.10934"
arxiv: "2004.10934"
licence: "arXiv — non-exclusive license to distribute"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/bochkovskiy-2020-yolov4-optimal-speed-accuracy-object/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/bochkovskiy-2020-yolov4-optimal-speed-accuracy-object/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2020](https://md.research.elamur.ai/free/en/papers/2020/) / YOLOv4: Optimal Speed and Accuracy of Object Detection

# YOLOv4: Optimal Speed and Accuracy of Object Detection

**Authors:** Alexey Bochkovskiy, Chien-Yao Wang and Hong-Yuan Mark Liao · **Year:** 2020 · **Venue:** arXiv:2004.10934 · **Licence:** [arXiv — non-exclusive license to distribute](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2004.10934) · [PDF](https://arxiv.org/pdf/2004.10934) · [DOI 10.48550/arxiv.2004.10934](https://doi.org/10.48550/arxiv.2004.10934) · [arXiv 2004.10934](https://arxiv.org/abs/2004.10934) · [OpenAlex](https://openalex.org/W3018757597)

**Citations:** 10,288 citations (OpenAlex, 2026-10-14) · [115](https://md.research.elamur.ai/free/en/papers/bochkovskiy-2020-yolov4-optimal-speed-accuracy-object/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The most accurate CNN object detectors are usually too slow for real-time use, and training them often requires many GPUs with large mini-batches. This work aims for a detector that runs in real time on an ordinary GPU and can be trained on one consumer card, such as a 1080 Ti or 2080 Ti. The authors also judge speed by actual parallel throughput instead of theoretical BFLOP counts. They split published improvements into two groups. The "bag of freebies" covers changes that affect only training, such as augmentation, label smoothing and IoU-family losses. The "bag of specials" covers cheap plug-in modules and post-processing steps. They test combinations from both groups on MS COCO and adapt CBN, PAN and SAM so they work well with single-GPU training. The final YOLOv4 uses a CSPDarknet-53 backbone together with Mosaic augmentation, CmBN, SAT, Mish, DropBlock and CIoU loss. It reaches 43.5% AP (65.7% AP50) on COCO at about 65 FPS on a Tesla V100. That is roughly twice the speed of EfficientDet at comparable accuracy, so strong detection becomes possible on modest hardware.

## Key points
- On MS COCO test-dev with a CSPDarknet-53 backbone on a V100 (batch=1, no TensorRT), input size trades accuracy for speed:
  - 416 px: 41.2% AP at 96 FPS
  - 512 px: 43.0% AP at 83 FPS
  - 608 px: 43.5% AP at 62 FPS
- Compared with YOLOv3, AP improves by about 10% and FPS by about 12%. YOLOv4 runs about twice as fast as EfficientDet at similar AP; for example, EfficientDet-D2 reaches 43.0% AP at 41.7 FPS.
- The paper describes a detector as four parts: input, backbone, neck and head. The neck collects features from different stages and can include SPP, PAN, FPN or BiFPN. The head can be one-stage or two-stage, and anchor-based or anchor-free.
- Components added or evaluated include WRC, CSP, CmBN, SAT, Mish activation, Mosaic augmentation, DropBlock and CIoU loss.
- The code is released in the Darknet framework at github.com/AlexeyAB/darknet.

## Interesting ideas
- **Freebies vs. specials taxonomy.** Tricks are sorted by where they cost something: only at training time, or with a small inference overhead. This gives ablation studies a clean structure and makes deployment trade-offs explicit. It could be used to organize experiments in any ML project, including our NLP models.
- **Measure real throughput, not FLOPs.** Models are optimized for FPS on actual GPUs rather than theoretical operation counts. Low FLOPs do not always mean fast inference, so benchmarks for production systems should time the real target hardware.
- **Design for one consumer GPU.** Methods such as batch normalization were reworked to suit small-batch training on a single card. This matters for teams with limited compute and is worth considering when we fine-tune models ourselves.
- **Treat the box as a whole in the loss.** IoU-based losses (GIoU, DIoU, CIoU) score the predicted box as one shape instead of four independent coordinates. This makes the loss scale-invariant and lets it reflect overlap, center distance and aspect ratio. The same idea applies to any spatial prediction, such as locating objects in map or satellite imagery.
- **Multi-image augmentation.** Mosaic augmentation follows earlier methods like MixUp and CutMix, which build one training sample from several images. This adds context variety cheaply. The same principle of mixing samples could help with small or imbalanced datasets in other domains.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Academia Sinica](https://md.research.elamur.ai/free/en/labs/academia-sinica/) — Research institute, Taipei, Taiwan · 13 works · 11,163 citations

📚 Shelves (3): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 3,486 works

## Cite

Cite as: Alexey Bochkovskiy, Chien-Yao Wang and Hong-Yuan Mark Liao. 2020. YOLOv4: Optimal Speed and Accuracy of Object Detection. arXiv:2004.10934.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{bochkovskiy2020yolov4,
  title = {{YOLOv4: Optimal Speed and Accuracy of Object Detection}},
  author = {Alexey Bochkovskiy, Chien-Yao Wang and Hong-Yuan Mark Liao},
  year = {2020},
  doi = {10.48550/arxiv.2004.10934},
  eprint = {2004.10934},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2004.10934}
}
```

- [← previous BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://md.research.elamur.ai/free/en/papers/lewis-2020-bart-denoising-sequence-sequence-pre/)
- [next → Unsupervised Cross-lingual Representation Learning at Scale](https://md.research.elamur.ai/free/en/papers/conneau-2020-unsupervised-cross-lingual-representation-learning/)
- [↑ back to papers of 2020](https://md.research.elamur.ai/free/en/papers/2020/)

## References

102 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Navaneeth Bodla, Bharat Singh, Rama Chellappa, and Larry S Davis. Soft-NMS–improving object detection with one line of code. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 5561–5569, 2017. 4 [find](https://scholar.google.com/scholar?q=Navaneeth+Bodla%2C+Bharat+Singh%2C+Rama+Chellappa%2C+and+Larry+S+Davis%2E+Soft%2DNMS%E2%80%93improving+object+detection+with+one+line+of+code%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICC)
2. Zhaowei Cai and Nuno Vasconcelos. Cascade R-CNN: Delving into high quality object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 6154–6162, 2018. 12 [find](https://scholar.google.com/scholar?q=Zhaowei+Cai+and+Nuno+Vasconcelos%2E+Cascade+R%2DCNN%3A+Delving+into+high+quality+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+pages+6154%E2%80%936162%2C+2)
3. Jiale Cao, Yanwei Pang, Jungong Han, and Xuelong Li. Hierarchical shot detector. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 9705–9714, 2019. 12 [find](https://scholar.google.com/scholar?q=Jiale+Cao%2C+Yanwei+Pang%2C+Jungong+Han%2C+and+Xuelong+Li%2E+Hierarchical+shot+detector%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+pages+9705%E2%80%939714%2C+2019%2E+12)
4. Ping Chao, Chao-Yang Kao, Yu-Shan Ruan, Chien-Hsiang Huang, and Youn-Long Lin. HarDNet: A low memory traffic network. Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2019. 13 [find](https://scholar.google.com/scholar?q=Ping+Chao%2C+Chao%2DYang+Kao%2C+Yu%2DShan+Ruan%2C+Chien%2DHsiang+Huang%2C+and+Youn%2DLong+Lin%2E+HarDNet%3A+A+low+memory+traffic+network%2E+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+2019%2E+)
5. Liang-Chieh Chen, George Papandreou, Iasonas Kokkinos, Kevin Murphy, and Alan L Yuille. DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 40(4):834–848, 2017. 2, 4 [find](https://scholar.google.com/scholar?q=Liang%2DChieh+Chen%2C+George+Papandreou%2C+Iasonas+Kokkinos%2C+Kevin+Murphy%2C+and+Alan+L+Yuille%2E+DeepLab%3A+Semantic+image+segmentation+with+deep+convolutional+nets%2C+atrous+convolution%2C+and+fully+connected+CRFs%2E)
6. Pengguang Chen. GridMask data augmentation. arXiv preprint arXiv:2001.04086, 2020. 3 — [arXiv:2001.04086](https://arxiv.org/abs/2001.04086) [find](https://scholar.google.com/scholar?q=Pengguang+Chen%2E+GridMask+data+augmentation%2E+arXiv+preprint+arXiv%3A2001%2E04086%2C+2020%2E+3+%E2%80%94+%5BarXiv%3A2001%2E04086%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2001%2E04086%29)
7. Yukang Chen, Tong Yang, Xiangyu Zhang, Gaofeng Meng, Xinyu Xiao, and Jian Sun. DetNAS: Backbone search for object detection. In Advances in Neural Information Processing Systems (NeurIPS), pages 6638–6648, 2019. 2 [find](https://scholar.google.com/scholar?q=Yukang+Chen%2C+Tong+Yang%2C+Xiangyu+Zhang%2C+Gaofeng+Meng%2C+Xinyu+Xiao%2C+and+Jian+Sun%2E+DetNAS%3A+Backbone+search+for+object+detection%2E+In+Advances+in+Neural+Information+Processing+Systems+%28NeurIPS%29%2C+pages+6638%E2%80%93)
8. Jiwoong Choi, Dayoung Chun, Hyun Kim, and Hyuk-Jae Lee. Gaussian YOLOv3: An accurate and fast object detector using localization uncertainty for autonomous driving. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 502–511, 2019. 7 [find](https://scholar.google.com/scholar?q=Jiwoong+Choi%2C+Dayoung+Chun%2C+Hyun+Kim%2C+and+Hyuk%2DJae+Lee%2E+Gaussian+YOLOv3%3A+An+accurate+and+fast+object+detector+using+localization+uncertainty+for+autonomous+driving%2E+In+Proceedings+of+the+IEEE+Internat)
9. Jifeng Dai, Yi Li, Kaiming He, and Jian Sun. R-FCN: Object detection via region-based fully convolutional networks. In Advances in Neural Information Processing Systems (NIPS), pages 379–387, 2016. 2 [find](https://scholar.google.com/scholar?q=Jifeng+Dai%2C+Yi+Li%2C+Kaiming+He%2C+and+Jian+Sun%2E+R%2DFCN%3A+Object+detection+via+region%2Dbased+fully+convolutional+networks%2E+In+Advances+in+Neural+Information+Processing+Systems+%28NIPS%29%2C+pages+379%E2%80%93387%2C+2016%2E+2)
10. Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. ImageNet: A large-scale hierarchical image database. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 248–255, 2009. 5 [find](https://scholar.google.com/scholar?q=Jia+Deng%2C+Wei+Dong%2C+Richard+Socher%2C+Li%2DJia+Li%2C+Kai+Li%2C+and+Li+Fei%2DFei%2E+ImageNet%3A+A+large%2Dscale+hierarchical+image+database%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recogni)
11. Terrance DeVries and Graham W Taylor. Improved regularization of convolutional neural networks with CutOut. arXiv preprint arXiv:1708.04552, 2017. 3 — [arXiv:1708.04552](https://arxiv.org/abs/1708.04552) [find](https://scholar.google.com/scholar?q=Terrance+DeVries+and+Graham+W+Taylor%2E+Improved+regularization+of+convolutional+neural+networks+with+CutOut%2E+arXiv+preprint+arXiv%3A1708%2E04552%2C+2017%2E+3+%E2%80%94+%5BarXiv%3A1708%2E04552%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1708%2E045)
12. Xianzhi Du, Tsung-Yi Lin, Pengchong Jin, Golnaz Ghiasi, Mingxing Tan, Yin Cui, Quoc V Le, and Xiaodan Song. SpineNet: Learning scale-permuted backbone for recognition and localization. arXiv preprint arXiv:1912.05027, 2019. 2 [find](https://scholar.google.com/scholar?q=Xianzhi+Du%2C+Tsung%2DYi+Lin%2C+Pengchong+Jin%2C+Golnaz+Ghiasi%2C+Mingxing+Tan%2C+Yin+Cui%2C+Quoc+V+Le%2C+and+Xiaodan+Song%2E+SpineNet%3A+Learning+scale%2Dpermuted+backbone+for+recognition+and+localization%2E+arXiv+preprint+)
13. Kaiwen Duan, Song Bai, Lingxi Xie, Honggang Qi, Qingming Huang, and Qi Tian. CenterNet: Keypoint triplets for object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 6569–6578, 2019. 2, 12 [find](https://scholar.google.com/scholar?q=Kaiwen+Duan%2C+Song+Bai%2C+Lingxi+Xie%2C+Honggang+Qi%2C+Qingming+Huang%2C+and+Qi+Tian%2E+CenterNet%3A+Keypoint+triplets+for+object+detection%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28)
14. Cheng-Yang Fu, Mykhailo Shvets, and Alexander C Berg. RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free. arXiv preprint arXiv:1901.03353, 2019. 12 — [arXiv:1901.03353](https://arxiv.org/abs/1901.03353) [find](https://scholar.google.com/scholar?q=Cheng%2DYang+Fu%2C+Mykhailo+Shvets%2C+and+Alexander+C+Berg%2E+RetinaMask%3A+Learning+to+predict+masks+improves+state%2Dof%2Dthe%2Dart+single%2Dshot+detection+for+free%2E+arXiv+preprint+arXiv%3A1901%2E03353%2C+2019%2E+12+%E2%80%94+%5BarXiv)
15. Robert Geirhos, Patricia Rubisch, Claudio Michaelis, Matthias Bethge, Felix A Wichmann, and Wieland Brendel. ImageNet-trained cnns are biased towards texture; increasing shape bias improves accuracy and robustness. In International Conference on Learning Representations (ICLR), 2019. 3 [find](https://scholar.google.com/scholar?q=Robert+Geirhos%2C+Patricia+Rubisch%2C+Claudio+Michaelis%2C+Matthias+Bethge%2C+Felix+A+Wichmann%2C+and+Wieland+Brendel%2E+ImageNet%2Dtrained+cnns+are+biased+towards+texture%3B+increasing+shape+bias+improves+accuracy+a)
16. Golnaz Ghiasi, Tsung-Yi Lin, and Quoc V Le. DropBlock: A regularization method for convolutional networks. In Advances in Neural Information Processing Systems (NIPS), pages 10727–10737, 2018. 3 [find](https://scholar.google.com/scholar?q=Golnaz+Ghiasi%2C+Tsung%2DYi+Lin%2C+and+Quoc+V+Le%2E+DropBlock%3A+A+regularization+method+for+convolutional+networks%2E+In+Advances+in+Neural+Information+Processing+Systems+%28NIPS%29%2C+pages+10727%E2%80%9310737%2C+2018%2E+3)
17. Golnaz Ghiasi, Tsung-Yi Lin, and Quoc V Le. NAS-FPN: Learning scalable feature pyramid architecture for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 7036– 7045, 2019. 2, 13 [find](https://scholar.google.com/scholar?q=Golnaz+Ghiasi%2C+Tsung%2DYi+Lin%2C+and+Quoc+V+Le%2E+NAS%2DFPN%3A+Learning+scalable+feature+pyramid+architecture+for+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recogniti)
18. Ross Girshick. Fast R-CNN. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 1440–1448, 2015. 2 [find](https://scholar.google.com/scholar?q=Ross+Girshick%2E+Fast+R%2DCNN%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+pages+1440%E2%80%931448%2C+2015%2E+2)
19. Ross Girshick, Jeff Donahue, Trevor Darrell, and Jitendra Malik. [Rich feature hierarchies for accurate object detection and semantic segmentation](https://md.research.elamur.ai/free/en/papers/girshick-2013-rich-feature-hierarchies-accurate-object/). In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 580–587, 2014. 2, 4
20. Jianyuan Guo, Kai Han, Yunhe Wang, Chao Zhang, Zhaohui Yang, Han Wu, Xinghao Chen, and Chang Xu. Hit-Detector: Hierarchical trinity architecture search for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 2 [find](https://scholar.google.com/scholar?q=Jianyuan+Guo%2C+Kai+Han%2C+Yunhe+Wang%2C+Chao+Zhang%2C+Zhaohui+Yang%2C+Han+Wu%2C+Xinghao+Chen%2C+and+Chang+Xu%2E+Hit%2DDetector%3A+Hierarchical+trinity+architecture+search+for+object+detection%2E+In+Proceedings+of+the+IEEE)
21. Kai Han, Yunhe Wang, Qi Tian, Jianyuan Guo, Chunjing Xu, and Chang Xu. GhostNet: More features from cheap operations. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 5 [find](https://scholar.google.com/scholar?q=Kai+Han%2C+Yunhe+Wang%2C+Qi+Tian%2C+Jianyuan+Guo%2C+Chunjing+Xu%2C+and+Chang+Xu%2E+GhostNet%3A+More+features+from+cheap+operations%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28)
22. Bharath Hariharan, Pablo Arbeláez, Ross Girshick, and Jitendra Malik. Hypercolumns for object segmentation and fine-grained localization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 447–456, 2015. 4 [find](https://scholar.google.com/scholar?q=Bharath+Hariharan%2C+Pablo+Arbela%CC%81ez%2C+Ross+Girshick%2C+and+Jitendra+Malik%2E+Hypercolumns+for+object+segmentation+and+fine%2Dgrained+localization%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+)
23. Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross Girshick. Mask R-CNN. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 2961–2969, 2017. 2 [find](https://scholar.google.com/scholar?q=Kaiming+He%2C+Georgia+Gkioxari%2C+Piotr+Dolla%CC%81r%2C+and+Ross+Girshick%2E+Mask+R%2DCNN%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+pages+2961%E2%80%932969%2C+2017%2E+2)
24. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 1026–1034, 2015. 4 [find](https://scholar.google.com/scholar?q=Kaiming+He%2C+Xiangyu+Zhang%2C+Shaoqing+Ren%2C+and+Jian+Sun%2E+Delving+deep+into+rectifiers%3A+Surpassing+human%2Dlevel+performance+on+ImageNet+classification%2E+In+Proceedings+of+the+IEEE+International+Conference+)
25. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 37(9):1904–1916, 2015. 2, 4, 7 [find](https://scholar.google.com/scholar?q=Kaiming+He%2C+Xiangyu+Zhang%2C+Shaoqing+Ren%2C+and+Jian+Sun%2E+Spatial+pyramid+pooling+in+deep+convolutional+networks+for+visual+recognition%2E+IEEE+Transactions+on+Pattern+Analysis+and+Machine+Intelligence+%28TP)
26. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. [Deep residual learning for image recognition](https://md.research.elamur.ai/free/en/papers/he-2015-deep-residual-learning-image-recognition/). In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 770–778, 2016. 2
27. Andrew Howard, Mark Sandler, Grace Chu, Liang-Chieh Chen, Bo Chen, Mingxing Tan, Weijun Wang, Yukun Zhu, Ruoming Pang, Vijay Vasudevan, et al. Searching for MobileNetV3. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2019. 2, 4 [find](https://scholar.google.com/scholar?q=Andrew+Howard%2C+Mark+Sandler%2C+Grace+Chu%2C+Liang%2DChieh+Chen%2C+Bo+Chen%2C+Mingxing+Tan%2C+Weijun+Wang%2C+Yukun+Zhu%2C+Ruoming+Pang%2C+Vijay+Vasudevan%2C+et+al%2E+Searching+for+MobileNetV3%2E+In+Proceedings+of+the+IEEE+Int)
28. Andrew G Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, and Hartwig Adam. [MobileNets: Efficient convolutional neural networks for mobile vision applications](https://md.research.elamur.ai/free/en/papers/howard-2017-mobilenets-efficient-convolutional-neural-networks/). arXiv preprint arXiv:1704.04861, 2017. 2, 4
29. Jie Hu, Li Shen, and Gang Sun. Squeeze-and-excitation networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 7132– 7141, 2018. 4 [find](https://scholar.google.com/scholar?q=Jie+Hu%2C+Li+Shen%2C+and+Gang+Sun%2E+Squeeze%2Dand%2Dexcitation+networks%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+pages+7132%E2%80%93+7141%2C+2018%2E+4)
30. Gao Huang, Zhuang Liu, Laurens Van Der Maaten, and Kilian Q Weinberger. Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 4700– 4708, 2017. 2 [find](https://scholar.google.com/scholar?q=Gao+Huang%2C+Zhuang+Liu%2C+Laurens+Van+Der+Maaten%2C+and+Kilian+Q+Weinberger%2E+Densely+connected+convolutional+networks%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR)
31. Forrest N Iandola, Song Han, Matthew W Moskewicz, Khalid Ashraf, William J Dally, and Kurt Keutzer. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and¡ 0.5 MB model size. arXiv preprint arXiv:1602.07360, 2016. 2 [find](https://scholar.google.com/scholar?q=Forrest+N+Iandola%2C+Song+Han%2C+Matthew+W+Moskewicz%2C+Khalid+Ashraf%2C+William+J+Dally%2C+and+Kurt+Keutzer%2E+SqueezeNet%3A+AlexNet%2Dlevel+accuracy+with+50x+fewer+parameters+and%C2%A1+0%2E5+MB+model+size%2E+arXiv+preprint+)
32. Sergey Ioffe and Christian Szegedy. Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167, 2015. 6 [find](https://scholar.google.com/scholar?q=Sergey+Ioffe+and+Christian+Szegedy%2E+Batch+normalization%3A+Accelerating+deep+network+training+by+reducing+internal+covariate+shift%2E+arXiv+preprint+arXiv%3A1502%2E03167%2C+2015%2E+6)
33. Md Amirul Islam, Shujon Naha, Mrigank Rochan, Neil Bruce, and Yang Wang. Label refinement network for coarse-to-fine semantic segmentation. arXiv preprint arXiv:1703.00551, 2017. 3 — [arXiv:1703.00551](https://arxiv.org/abs/1703.00551) [find](https://scholar.google.com/scholar?q=Md+Amirul+Islam%2C+Shujon+Naha%2C+Mrigank+Rochan%2C+Neil+Bruce%2C+and+Yang+Wang%2E+Label+refinement+network+for+coarse%2Dto%2Dfine+semantic+segmentation%2E+arXiv+preprint+arXiv%3A1703%2E00551%2C+2017%2E+3+%E2%80%94+%5BarXiv%3A1703%2E00551)
34. Seung-Wook Kim, Hyong-Keun Kook, Jee-Young Sun, Mun-Cheon Kang, and Sung-Jea Ko. Parallel feature pyramid network for object detection. In Proceedings of the European Conference on Computer Vision (ECCV), pages 234–250, 2018. 11 [find](https://scholar.google.com/scholar?q=Seung%2DWook+Kim%2C+Hyong%2DKeun+Kook%2C+Jee%2DYoung+Sun%2C+Mun%2DCheon+Kang%2C+and+Sung%2DJea+Ko%2E+Parallel+feature+pyramid+network+for+object+detection%2E+In+Proceedings+of+the+European+Conference+on+Computer+Vision+%28EC)
35. Günter Klambauer, Thomas Unterthiner, Andreas Mayr, and Sepp Hochreiter. Self-normalizing neural networks. In Advances in Neural Information Processing Systems (NIPS), pages 971–980, 2017. 4 [find](https://scholar.google.com/scholar?q=Gu%CC%88nter+Klambauer%2C+Thomas+Unterthiner%2C+Andreas+Mayr%2C+and+Sepp+Hochreiter%2E+Self%2Dnormalizing+neural+networks%2E+In+Advances+in+Neural+Information+Processing+Systems+%28NIPS%29%2C+pages+971%E2%80%93980%2C+2017%2E+4)
36. Gustav Larsson, Michael Maire, and Gregory Shakhnarovich. FractalNet: Ultra-deep neural networks without residuals. arXiv preprint arXiv:1605.07648, 2016. 6 — [arXiv:1605.07648](https://arxiv.org/abs/1605.07648) [find](https://scholar.google.com/scholar?q=Gustav+Larsson%2C+Michael+Maire%2C+and+Gregory+Shakhnarovich%2E+FractalNet%3A+Ultra%2Ddeep+neural+networks+without+residuals%2E+arXiv+preprint+arXiv%3A1605%2E07648%2C+2016%2E+6+%E2%80%94+%5BarXiv%3A1605%2E07648%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F)
37. Hei Law and Jia Deng. CornerNet: Detecting objects as paired keypoints. In Proceedings of the European Conference on Computer Vision (ECCV), pages 734–750, 2018. 2, 11 [find](https://scholar.google.com/scholar?q=Hei+Law+and+Jia+Deng%2E+CornerNet%3A+Detecting+objects+as+paired+keypoints%2E+In+Proceedings+of+the+European+Conference+on+Computer+Vision+%28ECCV%29%2C+pages+734%E2%80%93750%2C+2018%2E+2%2C+11)
38. Hei Law, Yun Teng, Olga Russakovsky, and Jia Deng. CornerNet-Lite: Efficient keypoint based object detection. arXiv preprint arXiv:1904.08900, 2019. 2 — [arXiv:1904.08900](https://arxiv.org/abs/1904.08900) [find](https://scholar.google.com/scholar?q=Hei+Law%2C+Yun+Teng%2C+Olga+Russakovsky%2C+and+Jia+Deng%2E+CornerNet%2DLite%3A+Efficient+keypoint+based+object+detection%2E+arXiv+preprint+arXiv%3A1904%2E08900%2C+2019%2E+2+%E2%80%94+%5BarXiv%3A1904%2E08900%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1904%2E0)
39. Svetlana Lazebnik, Cordelia Schmid, and Jean Ponce. Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), volume 2, pages 2169–2178. IEEE, 2006. 4 [find](https://scholar.google.com/scholar?q=Svetlana+Lazebnik%2C+Cordelia+Schmid%2C+and+Jean+Ponce%2E+Beyond+bags+of+features%3A+Spatial+pyramid+matching+for+recognizing+natural+scene+categories%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision)
40. Youngwan Lee and Jongyoul Park. CenterMask: Real-time anchor-free instance segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 12, 13 [find](https://scholar.google.com/scholar?q=Youngwan+Lee+and+Jongyoul+Park%2E+CenterMask%3A+Real%2Dtime+anchor%2Dfree+instance+segmentation%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+2020%2E+12%2C+13)
41. Shuai Li, Lingxiao Yang, Jianqiang Huang, Xian-Sheng Hua, and Lei Zhang. Dynamic anchor feature selection for single-shot object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 6609–6618, 2019. 12 [find](https://scholar.google.com/scholar?q=Shuai+Li%2C+Lingxiao+Yang%2C+Jianqiang+Huang%2C+Xian%2DSheng+Hua%2C+and+Lei+Zhang%2E+Dynamic+anchor+feature+selection+for+single%2Dshot+object+detection%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Compu)
42. Yanghao Li, Yuntao Chen, Naiyan Wang, and Zhaoxiang Zhang. Scale-aware trident networks for object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 6054–6063, 2019. 12 [find](https://scholar.google.com/scholar?q=Yanghao+Li%2C+Yuntao+Chen%2C+Naiyan+Wang%2C+and+Zhaoxiang+Zhang%2E+Scale%2Daware+trident+networks+for+object+detection%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+pages+6054%E2%80%93)
43. Zeming Li, Chao Peng, Gang Yu, Xiangyu Zhang, Yangdong Deng, and Jian Sun. DetNet: Design backbone for object detection. In Proceedings of the European Conference on Computer Vision (ECCV), pages 334–350, 2018. 2 [find](https://scholar.google.com/scholar?q=Zeming+Li%2C+Chao+Peng%2C+Gang+Yu%2C+Xiangyu+Zhang%2C+Yangdong+Deng%2C+and+Jian+Sun%2E+DetNet%3A+Design+backbone+for+object+detection%2E+In+Proceedings+of+the+European+Conference+on+Computer+Vision+%28ECCV%29%2C+pages+334%E2%80%93)
44. Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Bharath Hariharan, and Serge Belongie. [Feature pyramid networks for object detection](https://md.research.elamur.ai/free/en/papers/lin-2016-feature-pyramid-networks-object-detection/). In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2117–2125, 2017. 2
45. Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollár. [Focal loss for dense object detection](https://md.research.elamur.ai/free/en/papers/lin-2017-focal-loss-dense-object-detection/). In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 2980–2988, 2017. 2, 3, 11, 13
46. Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. [Microsoft COCO: Common objects in context](https://md.research.elamur.ai/free/en/papers/lin-2014-microsoft-coco-common-objects-context/). In Proceedings of the European Conference on Computer Vision (ECCV), pages 740–755, 2014. 5
47. Songtao Liu, Di Huang, et al. Receptive field block net for accurate and fast object detection. In Proceedings of the European Conference on Computer Vision (ECCV), pages 385–400, 2018. 2, 4, 11 [find](https://scholar.google.com/scholar?q=Songtao+Liu%2C+Di+Huang%2C+et+al%2E+Receptive+field+block+net+for+accurate+and+fast+object+detection%2E+In+Proceedings+of+the+European+Conference+on+Computer+Vision+%28ECCV%29%2C+pages+385%E2%80%93400%2C+2018%2E+2%2C+4%2C+11)
48. Songtao Liu, Di Huang, and Yunhong Wang. Learning spatial fusion for single-shot object detection. arXiv preprint arXiv:1911.09516, 2019. 2, 4, 13 — [arXiv:1911.09516](https://arxiv.org/abs/1911.09516) [find](https://scholar.google.com/scholar?q=Songtao+Liu%2C+Di+Huang%2C+and+Yunhong+Wang%2E+Learning+spatial+fusion+for+single%2Dshot+object+detection%2E+arXiv+preprint+arXiv%3A1911%2E09516%2C+2019%2E+2%2C+4%2C+13+%E2%80%94+%5BarXiv%3A1911%2E09516%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1911%2E09516)
49. Shu Liu, Lu Qi, Haifang Qin, Jianping Shi, and Jiaya Jia. Path aggregation network for instance segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 8759–8768, 2018. 1, 2, 7 [find](https://scholar.google.com/scholar?q=Shu+Liu%2C+Lu+Qi%2C+Haifang+Qin%2C+Jianping+Shi%2C+and+Jiaya+Jia%2E+Path+aggregation+network+for+instance+segmentation%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+p)
50. Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, and Alexander C Berg. SSD: Single shot multibox detector. In Proceedings of the European Conference on Computer Vision (ECCV), pages 21–37, 2016. 2, 11 [find](https://scholar.google.com/scholar?q=Wei+Liu%2C+Dragomir+Anguelov%2C+Dumitru+Erhan%2C+Christian+Szegedy%2C+Scott+Reed%2C+Cheng%2DYang+Fu%2C+and+Alexander+C+Berg%2E+SSD%3A+Single+shot+multibox+detector%2E+In+Proceedings+of+the+European+Conference+on+Computer)
51. Jonathan Long, Evan Shelhamer, and Trevor Darrell. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 3431–3440, 2015. 4 [find](https://scholar.google.com/scholar?q=Jonathan+Long%2C+Evan+Shelhamer%2C+and+Trevor+Darrell%2E+Fully+convolutional+networks+for+semantic+segmentation%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+page)
52. Ilya Loshchilov and Frank Hutter. SGDR: Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983, 2016. 7 [find](https://scholar.google.com/scholar?q=Ilya+Loshchilov+and+Frank+Hutter%2E+SGDR%3A+Stochastic+gradient+descent+with+warm+restarts%2E+arXiv+preprint+arXiv%3A1608%2E03983%2C+2016%2E+7)
53. Ningning Ma, Xiangyu Zhang, Hai-Tao Zheng, and Jian Sun. ShuffleNetV2: Practical guidelines for efficient cnn architecture design. In Proceedings of the European Conference on Computer Vision (ECCV), pages 116–131, 2018. 2 [find](https://scholar.google.com/scholar?q=Ningning+Ma%2C+Xiangyu+Zhang%2C+Hai%2DTao+Zheng%2C+and+Jian+Sun%2E+ShuffleNetV2%3A+Practical+guidelines+for+efficient+cnn+architecture+design%2E+In+Proceedings+of+the+European+Conference+on+Computer+Vision+%28ECCV%29%2C+)
54. Andrew L Maas, Awni Y Hannun, and Andrew Y Ng. Rectifier nonlinearities improve neural network acoustic models. In Proceedings of International Conference on Machine Learning (ICML), volume 30, page 3, 2013. 4 [find](https://scholar.google.com/scholar?q=Andrew+L+Maas%2C+Awni+Y+Hannun%2C+and+Andrew+Y+Ng%2E+Rectifier+nonlinearities+improve+neural+network+acoustic+models%2E+In+Proceedings+of+International+Conference+on+Machine+Learning+%28ICML%29%2C+volume+30%2C+page+3)
55. Diganta Misra. Mish: A self regularized non-monotonic neural activation function. arXiv preprint arXiv:1908.08681, 2019. 4 — [arXiv:1908.08681](https://arxiv.org/abs/1908.08681) [find](https://scholar.google.com/scholar?q=Diganta+Misra%2E+Mish%3A+A+self+regularized+non%2Dmonotonic+neural+activation+function%2E+arXiv+preprint+arXiv%3A1908%2E08681%2C+2019%2E+4+%E2%80%94+%5BarXiv%3A1908%2E08681%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1908%2E08681%29)
56. Vinod Nair and Geoffrey E Hinton. Rectified linear units improve restricted boltzmann machines. In Proceedings of International Conference on Machine Learning (ICML), pages 807–814, 2010. 4 [find](https://scholar.google.com/scholar?q=Vinod+Nair+and+Geoffrey+E+Hinton%2E+Rectified+linear+units+improve+restricted+boltzmann+machines%2E+In+Proceedings+of+International+Conference+on+Machine+Learning+%28ICML%29%2C+pages+807%E2%80%93814%2C+2010%2E+4)
57. Jing Nie, Rao Muhammad Anwer, Hisham Cholakkal, Fahad Shahbaz Khan, Yanwei Pang, and Ling Shao. Enriched feature guided refinement network for object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 9537–9546, 2019. 12 [find](https://scholar.google.com/scholar?q=Jing+Nie%2C+Rao+Muhammad+Anwer%2C+Hisham+Cholakkal%2C+Fahad+Shahbaz+Khan%2C+Yanwei+Pang%2C+and+Ling+Shao%2E+Enriched+feature+guided+refinement+network+for+object+detection%2E+In+Proceedings+of+the+IEEE+Internationa)
58. Jiangmiao Pang, Kai Chen, Jianping Shi, Huajun Feng, Wanli Ouyang, and Dahua Lin. Libra R-CNN: Towards balanced learning for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 821–830, 2019. 2, 12 [find](https://scholar.google.com/scholar?q=Jiangmiao+Pang%2C+Kai+Chen%2C+Jianping+Shi%2C+Huajun+Feng%2C+Wanli+Ouyang%2C+and+Dahua+Lin%2E+Libra+R%2DCNN%3A+Towards+balanced+learning+for+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+)
59. Prajit Ramachandran, Barret Zoph, and Quoc V Le. Searching for activation functions. arXiv preprint arXiv:1710.05941, 2017. 4 [find](https://scholar.google.com/scholar?q=Prajit+Ramachandran%2C+Barret+Zoph%2C+and+Quoc+V+Le%2E+Searching+for+activation+functions%2E+arXiv+preprint+arXiv%3A1710%2E05941%2C+2017%2E+4)
60. Abdullah Rashwan, Agastya Kalra, and Pascal Poupart. Matrix Nets: A new deep architecture for object detection. In Proceedings of the IEEE International Conference on Computer Vision Workshop (ICCV Workshop), pages 0–0, 2019. 2 [find](https://scholar.google.com/scholar?q=Abdullah+Rashwan%2C+Agastya+Kalra%2C+and+Pascal+Poupart%2E+Matrix+Nets%3A+A+new+deep+architecture+for+object+detection%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+Workshop+%28ICCV+Wo)
61. Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. You only look once: Unified, real-time object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 779– 788, 2016. 2 [find](https://scholar.google.com/scholar?q=Joseph+Redmon%2C+Santosh+Divvala%2C+Ross+Girshick%2C+and+Ali+Farhadi%2E+You+only+look+once%3A+Unified%2C+real%2Dtime+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognitio)
62. Joseph Redmon and Ali Farhadi. YOLO9000: better, faster, stronger. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 7263– 7271, 2017. 2 [find](https://scholar.google.com/scholar?q=Joseph+Redmon+and+Ali+Farhadi%2E+YOLO9000%3A+better%2C+faster%2C+stronger%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+pages+7263%E2%80%93+7271%2C+2017%2E+2)
63. Joseph Redmon and Ali Farhadi. YOLOv3: An incremental improvement. arXiv preprint arXiv:1804.02767, 2018. 2, 4, 7, 11 — [arXiv:1804.02767](https://arxiv.org/abs/1804.02767) [find](https://scholar.google.com/scholar?q=Joseph+Redmon+and+Ali+Farhadi%2E+YOLOv3%3A+An+incremental+improvement%2E+arXiv+preprint+arXiv%3A1804%2E02767%2C+2018%2E+2%2C+4%2C+7%2C+11+%E2%80%94+%5BarXiv%3A1804%2E02767%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1804%2E02767%29)
64. Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. [Faster R-CNN: Towards real-time object detection with region proposal networks](https://md.research.elamur.ai/free/en/papers/ren-2015-faster-r-cnn-towards-real/). In Advances in Neural Information Processing Systems (NIPS), pages 91–99, 2015. 2
65. Hamid Rezatofighi, Nathan Tsoi, JunYoung Gwak, Amir Sadeghian, Ian Reid, and Silvio Savarese. Generalized intersection over union: A metric and a loss for bounding box regression. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 658–666, 2019. 3 [find](https://scholar.google.com/scholar?q=Hamid+Rezatofighi%2C+Nathan+Tsoi%2C+JunYoung+Gwak%2C+Amir+Sadeghian%2C+Ian+Reid%2C+and+Silvio+Savarese%2E+Generalized+intersection+over+union%3A+A+metric+and+a+loss+for+bounding+box+regression%2E+In+Proceedings+of+th)
66. Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. MobileNetV2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 4510–4520, 2018. 2 [find](https://scholar.google.com/scholar?q=Mark+Sandler%2C+Andrew+Howard%2C+Menglong+Zhu%2C+Andrey+Zhmoginov%2C+and+Liang%2DChieh+Chen%2E+MobileNetV2%3A+Inverted+residuals+and+linear+bottlenecks%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+)
67. Abhinav Shrivastava, Abhinav Gupta, and Ross Girshick. Training region-based object detectors with online hard example mining. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 761–769, 2016. 3 [find](https://scholar.google.com/scholar?q=Abhinav+Shrivastava%2C+Abhinav+Gupta%2C+and+Ross+Girshick%2E+Training+region%2Dbased+object+detectors+with+online+hard+example+mining%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Reco)
68. Karen Simonyan and Andrew Zisserman. [Very deep convolutional networks for large-scale image recognition](https://md.research.elamur.ai/free/en/papers/simonyan-2014-very-deep-convolutional-networks-large/). arXiv preprint arXiv:1409.1556, 2014. 2
69. Krishna Kumar Singh, Hao Yu, Aron Sarmasi, Gautam Pradeep, and Yong Jae Lee. Hide-and-Seek: A data augmentation technique for weakly-supervised localization and beyond. arXiv preprint arXiv:1811.02545, 2018. 3 — [arXiv:1811.02545](https://arxiv.org/abs/1811.02545) [find](https://scholar.google.com/scholar?q=Krishna+Kumar+Singh%2C+Hao+Yu%2C+Aron+Sarmasi%2C+Gautam+Pradeep%2C+and+Yong+Jae+Lee%2E+Hide%2Dand%2DSeek%3A+A+data+augmentation+technique+for+weakly%2Dsupervised+localization+and+beyond%2E+arXiv+preprint+arXiv%3A1811%2E02545)
70. Saurabh Singh and Shankar Krishnan. Filter response normalization layer: Eliminating batch dependence in the training of deep neural networks. arXiv preprint arXiv:1911.09737, 2019. 6 — [arXiv:1911.09737](https://arxiv.org/abs/1911.09737) [find](https://scholar.google.com/scholar?q=Saurabh+Singh+and+Shankar+Krishnan%2E+Filter+response+normalization+layer%3A+Eliminating+batch+dependence+in+the+training+of+deep+neural+networks%2E+arXiv+preprint+arXiv%3A1911%2E09737%2C+2019%2E+6+%E2%80%94+%5BarXiv%3A1911%2E09)
71. Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov. [DropOut: A simple way to prevent neural networks from overfitting](https://md.research.elamur.ai/free/en/papers/srivastava-2014-dropout-simple-way-prevent/). The journal of machine learning research, 15(1):1929–1958, 2014. 3
72. K-K Sung and Tomaso Poggio. Example-based learning for view-based human face detection. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 20(1):39–51, 1998. 3 [find](https://scholar.google.com/scholar?q=K%2DK+Sung+and+Tomaso+Poggio%2E+Example%2Dbased+learning+for+view%2Dbased+human+face+detection%2E+IEEE+Transactions+on+Pattern+Analysis+and+Machine+Intelligence+%28TPAMI%29%2C+20%281%29%3A39%E2%80%9351%2C+1998%2E+3)
73. Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jon Shlens, and Zbigniew Wojna. Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2818–2826, 2016. 3 [find](https://scholar.google.com/scholar?q=Christian+Szegedy%2C+Vincent+Vanhoucke%2C+Sergey+Ioffe%2C+Jon+Shlens%2C+and+Zbigniew+Wojna%2E+Rethinking+the+inception+architecture+for+computer+vision%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+)
74. Mingxing Tan, Bo Chen, Ruoming Pang, Vijay Vasudevan, Mark Sandler, Andrew Howard, and Quoc V Le. MNASnet: Platform-aware neural architecture search for mobile. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2820–2828, 2019. 2 [find](https://scholar.google.com/scholar?q=Mingxing+Tan%2C+Bo+Chen%2C+Ruoming+Pang%2C+Vijay+Vasudevan%2C+Mark+Sandler%2C+Andrew+Howard%2C+and+Quoc+V+Le%2E+MNASnet%3A+Platform%2Daware+neural+architecture+search+for+mobile%2E+In+Proceedings+of+the+IEEE+Conference+o)
75. Mingxing Tan and Quoc V Le. [EfficientNet: Rethinking model scaling for convolutional neural networks](https://md.research.elamur.ai/free/en/papers/tan-2019-efficientnet-rethinking-model-scaling-convolutional/). In Proceedings of International Conference on Machine Learning (ICML), 2019. 2
76. Mingxing Tan and Quoc V Le. MixNet: Mixed depthwise convolutional kernels. In Proceedings of the British Machine Vision Conference (BMVC), 2019. 5 [find](https://scholar.google.com/scholar?q=Mingxing+Tan+and+Quoc+V+Le%2E+MixNet%3A+Mixed+depthwise+convolutional+kernels%2E+In+Proceedings+of+the+British+Machine+Vision+Conference+%28BMVC%29%2C+2019%2E+5)
77. Mingxing Tan, Ruoming Pang, and Quoc V Le. Efficient-Det: Scalable and efficient object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 2, 4, 13 [find](https://scholar.google.com/scholar?q=Mingxing+Tan%2C+Ruoming+Pang%2C+and+Quoc+V+Le%2E+Efficient%2DDet%3A+Scalable+and+efficient+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+2020%2E+2%2C+4%2C+)
78. Zhi Tian, Chunhua Shen, Hao Chen, and Tong He. FCOS: Fully convolutional one-stage object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 9627–9636, 2019. 2 [find](https://scholar.google.com/scholar?q=Zhi+Tian%2C+Chunhua+Shen%2C+Hao+Chen%2C+and+Tong+He%2E+FCOS%3A+Fully+convolutional+one%2Dstage+object+detection%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+pages+9627%E2%80%939636%2C+201)
79. Jonathan Tompson, Ross Goroshin, Arjun Jain, Yann Le-Cun, and Christoph Bregler. Efficient object localization using convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 648–656, 2015. 6 [find](https://scholar.google.com/scholar?q=Jonathan+Tompson%2C+Ross+Goroshin%2C+Arjun+Jain%2C+Yann+Le%2DCun%2C+and+Christoph+Bregler%2E+Efficient+object+localization+using+convolutional+networks%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+an)
80. Li Wan, Matthew Zeiler, Sixin Zhang, Yann Le Cun, and Rob Fergus. Regularization of neural networks using Drop-Connect. In Proceedings of International Conference on Machine Learning (ICML), pages 1058–1066, 2013. 3 [find](https://scholar.google.com/scholar?q=Li+Wan%2C+Matthew+Zeiler%2C+Sixin+Zhang%2C+Yann+Le+Cun%2C+and+Rob+Fergus%2E+Regularization+of+neural+networks+using+Drop%2DConnect%2E+In+Proceedings+of+International+Conference+on+Machine+Learning+%28ICML%29%2C+pages+105)
81. Chien-Yao Wang, Hong-Yuan Mark Liao, Yueh-Hua Wu, Ping-Yang Chen, Jun-Wei Hsieh, and I-Hau Yeh. CSPNet: A new backbone that can enhance learning capability of cnn. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshop (CVPR Workshop), 2020. 2, 7 [find](https://scholar.google.com/scholar?q=Chien%2DYao+Wang%2C+Hong%2DYuan+Mark+Liao%2C+Yueh%2DHua+Wu%2C+Ping%2DYang+Chen%2C+Jun%2DWei+Hsieh%2C+and+I%2DHau+Yeh%2E+CSPNet%3A+A+new+backbone+that+can+enhance+learning+capability+of+cnn%2E+Proceedings+of+the+IEEE+Conference+o)
82. Jiaqi Wang, Kai Chen, Shuo Yang, Chen Change Loy, and Dahua Lin. Region proposal by guided anchoring. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 2965–2974, 2019. 12 [find](https://scholar.google.com/scholar?q=Jiaqi+Wang%2C+Kai+Chen%2C+Shuo+Yang%2C+Chen+Change+Loy%2C+and+Dahua+Lin%2E+Region+proposal+by+guided+anchoring%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CVPR%29%2C+pages+296)
83. Shaoru Wang, Yongchao Gong, Junliang Xing, Lichao Huang, Chang Huang, and Weiming Hu. RDSNet: A new deep architecture for reciprocal object detection and instance segmentation. arXiv preprint arXiv:1912.05070, 2019. 13 — [arXiv:1912.05070](https://arxiv.org/abs/1912.05070) [find](https://scholar.google.com/scholar?q=Shaoru+Wang%2C+Yongchao+Gong%2C+Junliang+Xing%2C+Lichao+Huang%2C+Chang+Huang%2C+and+Weiming+Hu%2E+RDSNet%3A+A+new+deep+architecture+for+reciprocal+object+detection+and+instance+segmentation%2E+arXiv+preprint+arXiv%3A19)
84. Tiancai Wang, Rao Muhammad Anwer, Hisham Cholakkal, Fahad Shahbaz Khan, Yanwei Pang, and Ling Shao. Learning rich features at high-speed for single-shot object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 1971–1980, 2019. 11 [find](https://scholar.google.com/scholar?q=Tiancai+Wang%2C+Rao+Muhammad+Anwer%2C+Hisham+Cholakkal%2C+Fahad+Shahbaz+Khan%2C+Yanwei+Pang%2C+and+Ling+Shao%2E+Learning+rich+features+at+high%2Dspeed+for+single%2Dshot+object+detection%2E+In+Proceedings+of+the+IEEE+In)
85. Sanghyun Woo, Jongchan Park, Joon-Young Lee, and In So Kweon. CBAM: Convolutional block attention module. In Proceedings of the European Conference on Computer Vision (ECCV), pages 3–19, 2018. 1, 2, 4 [find](https://scholar.google.com/scholar?q=Sanghyun+Woo%2C+Jongchan+Park%2C+Joon%2DYoung+Lee%2C+and+In+So+Kweon%2E+CBAM%3A+Convolutional+block+attention+module%2E+In+Proceedings+of+the+European+Conference+on+Computer+Vision+%28ECCV%29%2C+pages+3%E2%80%9319%2C+2018%2E+1%2C+2%2C+4)
86. Saining Xie, Ross Girshick, Piotr Dollár, Zhuowen Tu, and Kaiming He. Aggregated residual transformations for deep neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 1492–1500, 2017. 2 [find](https://scholar.google.com/scholar?q=Saining+Xie%2C+Ross+Girshick%2C+Piotr+Dolla%CC%81r%2C+Zhuowen+Tu%2C+and+Kaiming+He%2E+Aggregated+residual+transformations+for+deep+neural+networks%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Patter)
87. Ze Yang, Shaohui Liu, Han Hu, Liwei Wang, and Stephen Lin. RepPoints: Point set representation for object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 9657–9666, 2019. 2, 12 [find](https://scholar.google.com/scholar?q=Ze+Yang%2C+Shaohui+Liu%2C+Han+Hu%2C+Liwei+Wang%2C+and+Stephen+Lin%2E+RepPoints%3A+Point+set+representation+for+object+detection%2E+In+Proceedings+of+the+IEEE+International+Conference+on+Computer+Vision+%28ICCV%29%2C+page)
88. Lewei Yao, Hang Xu, Wei Zhang, Xiaodan Liang, and Zhenguo Li. SM-NAS: Structural-to-modular neural architecture search for object detection. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2020. 13 [find](https://scholar.google.com/scholar?q=Lewei+Yao%2C+Hang+Xu%2C+Wei+Zhang%2C+Xiaodan+Liang%2C+and+Zhenguo+Li%2E+SM%2DNAS%3A+Structural%2Dto%2Dmodular+neural+architecture+search+for+object+detection%2E+In+Proceedings+of+the+AAAI+Conference+on+Artificial+Intelli)
89. Zhuliang Yao, Yue Cao, Shuxin Zheng, Gao Huang, and Stephen Lin. Cross-iteration batch normalization. arXiv preprint arXiv:2002.05712, 2020. 1, 6 — [arXiv:2002.05712](https://arxiv.org/abs/2002.05712) [find](https://scholar.google.com/scholar?q=Zhuliang+Yao%2C+Yue+Cao%2C+Shuxin+Zheng%2C+Gao+Huang%2C+and+Stephen+Lin%2E+Cross%2Diteration+batch+normalization%2E+arXiv+preprint+arXiv%3A2002%2E05712%2C+2020%2E+1%2C+6+%E2%80%94+%5BarXiv%3A2002%2E05712%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2002%2E05712%29)
90. Jiahui Yu, Yuning Jiang, Zhangyang Wang, Zhimin Cao, and Thomas Huang. UnitBox: An advanced object detection network. In Proceedings of the 24th ACM international conference on Multimedia, pages 516–520, 2016. 3 [find](https://scholar.google.com/scholar?q=Jiahui+Yu%2C+Yuning+Jiang%2C+Zhangyang+Wang%2C+Zhimin+Cao%2C+and+Thomas+Huang%2E+UnitBox%3A+An+advanced+object+detection+network%2E+In+Proceedings+of+the+24th+ACM+international+conference+on+Multimedia%2C+pages+516%E2%80%935)
91. Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo. CutMix: Regularization strategy to train strong classifiers with localizable features. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 6023–6032, 2019. 3 [find](https://scholar.google.com/scholar?q=Sangdoo+Yun%2C+Dongyoon+Han%2C+Seong+Joon+Oh%2C+Sanghyuk+Chun%2C+Junsuk+Choe%2C+and+Youngjoon+Yoo%2E+CutMix%3A+Regularization+strategy+to+train+strong+classifiers+with+localizable+features%2E+In+Proceedings+of+the+IE)
92. Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. [MixUp: Beyond empirical risk minimization](https://md.research.elamur.ai/free/en/papers/zhang-2017-mixup-empirical-risk-minimization/). arXiv preprint arXiv:1710.09412, 2017. 3
93. Hang Zhang, Kristin Dana, Jianping Shi, Zhongyue Zhang, Xiaogang Wang, Ambrish Tyagi, and Amit Agrawal. Context encoding for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 7151–7160, 2018. 6 [find](https://scholar.google.com/scholar?q=Hang+Zhang%2C+Kristin+Dana%2C+Jianping+Shi%2C+Zhongyue+Zhang%2C+Xiaogang+Wang%2C+Ambrish+Tyagi%2C+and+Amit+Agrawal%2E+Context+encoding+for+semantic+segmentation%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vi)
94. Shifeng Zhang, Cheng Chi, Yongqiang Yao, Zhen Lei, and Stan Z Li. Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. 13 [find](https://scholar.google.com/scholar?q=Shifeng+Zhang%2C+Cheng+Chi%2C+Yongqiang+Yao%2C+Zhen+Lei%2C+and+Stan+Z+Li%2E+Bridging+the+gap+between+anchor%2Dbased+and+anchor%2Dfree+detection+via+adaptive+training+sample+selection%2E+In+Proceedings+of+the+IEEE+Con)
95. Shifeng Zhang, Longyin Wen, Xiao Bian, Zhen Lei, and Stan Z Li. Single-shot refinement neural network for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 4203–4212, 2018. 11 [find](https://scholar.google.com/scholar?q=Shifeng+Zhang%2C+Longyin+Wen%2C+Xiao+Bian%2C+Zhen+Lei%2C+and+Stan+Z+Li%2E+Single%2Dshot+refinement+neural+network+for+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recogni)
96. Xiaosong Zhang, Fang Wan, Chang Liu, Rongrong Ji, and Qixiang Ye. FreeAnchor: Learning to match anchors for visual object detection. In Advances in Neural Information Processing Systems (NeurIPS), 2019. 12 [find](https://scholar.google.com/scholar?q=Xiaosong+Zhang%2C+Fang+Wan%2C+Chang+Liu%2C+Rongrong+Ji%2C+and+Qixiang+Ye%2E+FreeAnchor%3A+Learning+to+match+anchors+for+visual+object+detection%2E+In+Advances+in+Neural+Information+Processing+Systems+%28NeurIPS%29%2C+201)
97. Xiangyu Zhang, Xinyu Zhou, Mengxiao Lin, and Jian Sun. ShuffleNet: An extremely efficient convolutional neural network for mobile devices. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 6848–6856, 2018. 2 [find](https://scholar.google.com/scholar?q=Xiangyu+Zhang%2C+Xinyu+Zhou%2C+Mengxiao+Lin%2C+and+Jian+Sun%2E+ShuffleNet%3A+An+extremely+efficient+convolutional+neural+network+for+mobile+devices%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+)
98. Qijie Zhao, Tao Sheng, Yongtao Wang, Zhi Tang, Ying Chen, Ling Cai, and Haibin Ling. M2det: A single-shot object detector based on multi-level feature pyramid network. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), volume 33, pages 9259–9266, 2019. 2, 4, 11 [find](https://scholar.google.com/scholar?q=Qijie+Zhao%2C+Tao+Sheng%2C+Yongtao+Wang%2C+Zhi+Tang%2C+Ying+Chen%2C+Ling+Cai%2C+and+Haibin+Ling%2E+M2det%3A+A+single%2Dshot+object+detector+based+on+multi%2Dlevel+feature+pyramid+network%2E+In+Proceedings+of+the+AAAI+Confe)
99. Zhaohui Zheng, Ping Wang, Wei Liu, Jinze Li, Rongguang Ye, and Dongwei Ren. Distance-IoU Loss: Faster and better learning for bounding box regression. In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2020. 3, 4 [find](https://scholar.google.com/scholar?q=Zhaohui+Zheng%2C+Ping+Wang%2C+Wei+Liu%2C+Jinze+Li%2C+Rongguang+Ye%2C+and+Dongwei+Ren%2E+Distance%2DIoU+Loss%3A+Faster+and+better+learning+for+bounding+box+regression%2E+In+Proceedings+of+the+AAAI+Conference+on+Artifici)
100. Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang. Random erasing data augmentation. arXiv preprint arXiv:1708.04896, 2017. 3 [find](https://scholar.google.com/scholar?q=Zhun+Zhong%2C+Liang+Zheng%2C+Guoliang+Kang%2C+Shaozi+Li%2C+and+Yi+Yang%2E+Random+erasing+data+augmentation%2E+arXiv+preprint+arXiv%3A1708%2E04896%2C+2017%2E+3)
101. Chenchen Zhu, Fangyi Chen, Zhiqiang Shen, and Marios Savvides. Soft anchor-point object detection. arXiv preprint arXiv:1911.12448, 2019. 12 — [arXiv:1911.12448](https://arxiv.org/abs/1911.12448) [find](https://scholar.google.com/scholar?q=Chenchen+Zhu%2C+Fangyi+Chen%2C+Zhiqiang+Shen%2C+and+Marios+Savvides%2E+Soft+anchor%2Dpoint+object+detection%2E+arXiv+preprint+arXiv%3A1911%2E12448%2C+2019%2E+12+%E2%80%94+%5BarXiv%3A1911%2E12448%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F1911%2E12448%29)
102. Chenchen Zhu, Yihui He, and Marios Savvides. Feature selective anchor-free module for single-shot object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 840–849, 2019. 11 [find](https://scholar.google.com/scholar?q=Chenchen+Zhu%2C+Yihui+He%2C+and+Marios+Savvides%2E+Feature+selective+anchor%2Dfree+module+for+single%2Dshot+object+detection%2E+In+Proceedings+of+the+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition+%28CV)

*Updated 2 November 2026*