Inferior and Coordinate Distillation for Object Detectors
Abstract
:1. Introduction
- We demonstrate that inferior distillation facilitates the learning of the student model. If the inferior layer is used to guide the learning, it will result in certain improvement.
- We propose inferior and coordinate distillation for object detection models, which enables the student to simultaneously focus on the inferior information and the information contained in the coordinate layer of the teacher model. This indicates that multi-level information is utilized to guide the learning of the student model.
- Experiments conducted on various detectors, including one-stage, two-stage, and anchor-free detectors verified the effectiveness of the proposed method.
2. Related Works
2.1. Object Detection
2.2. Knowledge Distillation
3. Materials and Methods
3.1. Refine Module
3.2. Overall Loss
4. Results and Discussion
4.1. Dataset and Measurement
4.2. Details
4.3. Main Results
4.4. Ablation Study
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Howard, A.; Sandler, M.; Chu, G.; Chen, L.-C.; Chen, B.; Tan, M.; Wang, W.; Zhu, Y.; Pang, R.; Vasudevan, V.; et al. Searching for MobileNetV3. In Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Korea, 27 October–29 November 2019. [Google Scholar]
- de Almeida, I.D.P.D.; Corriça, J.V.D.P.; Costa, A.P.D.A.; Costa, I.P.D.A.; Maêda, S.M.D.N.; Gomes, C.F.S.; Santos, M.D. Study of the Location of a Second Fleet for the Brazilian Navy: Structuring and Mathematical Modeling Using SAPEVO-M and VIKOR Methods. In Proceedings of the International Conference of Production Research–Americas, Bahía Blanca, Argentina, 9–11 December 2020; Rossit, D.A., Tohmé, F., Mejía Delgadillo, G., Eds.; Springer International Publishing: Cham, Switzerland, 2020; pp. 113–124. [Google Scholar]
- Hinton, G.E.; Vinyals, O.; Dean, J. Distilling the Knowledge in a Neural Network. arXiv 2015, arXiv:1503.02531. [Google Scholar]
- Romero, A.; Ballas, N.; Kahou, S.E.; Chassang, A.; Gatta, C.; Bengio, Y. FitNets: Hints for Thin Deep Nets. In Proceedings of the 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, 7–9 May 2015. [Google Scholar]
- Ji, M.; Heo, B.; Park, S. Show, Attend and Distill: Knowledge Distillation via Attention-Based Feature Matching. In Proceedings of the Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, Thirty-Third Conference on Innovative Applications of Artificial Intelligence, IAAI 2021, The Eleventh Symposium on Educational Advances in Artificial Intelligence, EAAI 2021, Virtual, 2–9 February 2021; pp. 7945–7952. [Google Scholar]
- Chen, P.; Liu, S.; Zhao, H.; Jia, J. Distilling Knowledge via Knowledge Review. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, Virtual, 19–25 June 2021; pp. 5008–5017. [Google Scholar]
- Zhao, B.; Cui, Q.; Song, R.; Qiu, Y.; Liang, J. Decoupled Knowledge Distillation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, Louisiana, 21–24 June 2022. [Google Scholar] [CrossRef]
- Song, J.; Chen, Y.; Ye, J.; Song, M. Spot-Adaptive Knowledge Distillation. IEEE Trans. Image Process. 2022, 31, 3359–3370. [Google Scholar] [CrossRef] [PubMed]
- Chen, G.; Choi, W.; Yu, X.; Han, T.X.; Chandraker, M. Learning Efficient Object Detection Models with Knowledge Distillation. In Proceedings of the Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems, Long Beach, CA, USA, 4–9 December 2017; pp. 742–751. [Google Scholar]
- Wang, T.; Yuan, L.; Zhang, X.; Feng, J. Distilling Object Detectors With Fine-Grained Feature Imitation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2019, Long Beach, CA, USA, 16–20 June 2019; pp. 4933–4942. [Google Scholar]
- Sun, R.; Tang, F.; Zhang, X.; Xiong, H.; Tian, Q. Distilling Object Detectors with Task Adaptive Regularization. arXiv 2020, arXiv:2006.13108. [Google Scholar]
- Guo, J.; Han, K.; Wang, Y.; Wu, H.; Chen, X.; Xu, C.; Xu, C. Distilling Object Detectors via Decoupled Features. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, Virtual, 19–25 June 2021; pp. 2154–2164. [Google Scholar]
- Lin, T.-Y.; Goyal, P.; Girshick, R.B.; He, K.; Dollár, P. Focal Loss for Dense Object Detection. IEEE Trans. Pattern Anal. Mach. Intell. 2020, 42, 318–327. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Tian, Z.; Shen, C.; Chen, H.; He, T. FCOS: Fully Convolutional One-Stage Object Detection. In Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, ICCV 2019, Seoul, Korea, 27 October–2 November 2019; pp. 9626–9635. [Google Scholar]
- Ge, Z.; Liu, S.; Wang, F.; Li, Z.; Sun, J. YOLOX: Exceeding YOLO Series in 2021. arXiv 2021, arXiv:2107.08430. [Google Scholar]
- Ren, S.; He, K.; Girshick, R.B.; Sun, J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans. Pattern Anal. Mach. Intell. 2017, 39, 1137–1149. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Yang, Z.; Li, Z.; Jiang, X.; Gong, Y.; Yuan, Z.; Zhao, D.; Yuan, C. Focal and Global Knowledge Distillation for Detectors. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, Louisiana, 21–24 June 2022. [Google Scholar]
- Li, G.; Li, X.; Wang, Y.; Zhang, S.; Wu, Y.; Liang, D. Knowledge Distillation for Object Detection via Rank Mimicking and Prediction-Guided Feature Imitation. In Proceedings of the Thirty-Sixth AAAI Conference on Artificial Intelligence, AAAI 2022, Thirty-Fourth Conference on Innovative Applications of Artificial Intelligence, IAAI 2022, The Twelveth Symposium on Educational Advances in Artificial Intelligence, EAAI 2022, Virtual, 22 February–1 March 2022; pp. 1306–1313. [Google Scholar]
- Shu, C.; Liu, Y.; Gao, J.; Yan, Z.; Shen, C. Channel-Wise Knowledge Distillation for Dense Prediction. In Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision, ICCV 2021, Montreal, QC, Canada, 10–17 October 2021; pp. 5291–5300. [Google Scholar]
- Woo, S.; Park, J.; Lee, J.-Y.; Kweon, I.S. CBAM: Convolutional Block Attention Module. In Proceedings of the Computer Vision—ECCV 2018—15th European Conference, Munich, Germany, 8–14 September 2018; pp. 3–19. [Google Scholar]
- Lin, T.-Y.; Maire, M.; Belongie, S.J.; Hays, J.; Perona, P.; Ramanan, D.; Dollár, P.; Zitnick, C.L. Microsoft COCO: Common Objects in Context. In Proceedings of the Computer Vision—ECCV 2014—13th European Conference, Zurich, Switzerland, 6–12 September 2014; pp. 740–755. [Google Scholar]
- Chen, K.; Wang, J.; Pang, J.; Cao, Y.; Xiong, Y.; Li, X.; Sun, S.; Feng, W.; Liu, Z.; Xu, J.; et al. MMDetection: Open MMLab Detection Toolbox and Benchmark. arXiv 2019, arXiv:1906.07155. [Google Scholar]
- Dai, X.; Jiang, Z.; Wu, Z.; Bao, Y.; Wang, Z.; Liu, S.; Zhou, E. General Instance Distillation for Object Detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2021, Virtual, 19–25 June 2021; pp. 7842–7851. [Google Scholar]
Method | mAP/% |
---|---|
Faster R-CNN_ResNet50 (S) | 38.4 |
Faster R-CNN_ResNet101 (T) | 39.8 |
FGFI | 39.3 (+0.9) |
GID | 40.2 (+1.8) |
RMFPI | 40.5 (+2.1) |
FGD | 40.4 (+2.0) |
Ours | 40.5 (+2.1) |
RetinaNet_ResNet50 (S) | 37.4 |
RetinaNet_ResNet101 (T) | 38.9 |
FGFI | 38.6 (+1.2) |
GID | 39.1 (+1.7) |
RMFPI | 39.6 (+2.2) |
FGD | 39.6 (+2.2) |
Ours | 39.8 (+2.4) |
FCOS_ResNet50 (S) | 38.5 |
FCOS_ResNet101 (T) | 40.8 |
GID | 42.0 (+4.5) |
RMFPI | 42.3 (+4.8) |
FGD | 42.7 (+4.2) |
Ours | 42.8 (+4.3) |
Detectors | FCOS | RetinaNet | |
---|---|---|---|
Teacher | |||
ResNet-50 | 38.5 | 37.4 | |
ResNet-101 | 42.8 | 39.8 | |
ResNext-101 | 42.5 | 40.2 |
Method | Coordinate Distillation | Inferior Distillation | Refine Module | mAP/% |
---|---|---|---|---|
FCOS-ResNet101Distill FCOS-ResNet50 | 38.5 | |||
√ | 40.1 (+1.6) | |||
√ | √ | 41.7 (+3.2) | ||
√ | √ | √ | 42.8(+4.3) |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, Y.; Li, Y.; Pan, Z. Inferior and Coordinate Distillation for Object Detectors. Sensors 2022, 22, 5719. https://doi.org/10.3390/s22155719
Zhang Y, Li Y, Pan Z. Inferior and Coordinate Distillation for Object Detectors. Sensors. 2022; 22(15):5719. https://doi.org/10.3390/s22155719
Chicago/Turabian StyleZhang, Yao, Yang Li, and Zhisong Pan. 2022. "Inferior and Coordinate Distillation for Object Detectors" Sensors 22, no. 15: 5719. https://doi.org/10.3390/s22155719
APA StyleZhang, Y., Li, Y., & Pan, Z. (2022). Inferior and Coordinate Distillation for Object Detectors. Sensors, 22(15), 5719. https://doi.org/10.3390/s22155719