High accuracy object detection via bounding box regression network

Abstract

As one of the primary computer vision problems, object detection aims to find and locate semantic objects in digital images. Different with object classification, which only recognizes an object to a certain class, object detection also needs to extract accurate locations of objects. In the state-of-the-art object detection algorithms, bounding box regression plays a critical role in order to achieve high localization accuracy. Almost all the popular deep learning based object detection algorithms have utilized bounding box regression for fine tuning of object locations. However, while bounding box regression is widely used, there is few study focused on the underlying rationale, performance dependencies, and performance evaluation. In this paper, we proposed a dedicated deep neural network for bounding box regression, and presented several methods to improve its performance. Some ad hoc experiments are conducted to prove the effectiveness of the network. Also, we apply the network as an auxiliary module to the faster R-CNN algorithm and test them on some real-world images. Experiment results show certain performance improvements on detection accuracy in term of mean IOU.

Keywords

deep learning / object detection / bounding box regression / IOU distribution

References

[1] Mikolajczyk K, Schmid C. An affine invariant interest point detector. Vancouver, Canada. In: Proceedings of European Conference on Computer Vision. Beilin: Springer, 2002, 128–142

[2] Dalal N, Triggs B. Histograms of oriented gradients for human detection. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. San Diego: IEEE, 2005, 886–893

[3] Lowe D G. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 2004, 60(2): 91–110

[4] Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M S, Berg A C, Li F F. ImageNet large scale visual recognition challenge. International Journal of Computer Vision, 2015, 115(3): 211–252

[5] He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2015, 770–778

[6] Han J, Zhang D, Cheng G, Liu N, Xu D. Advanced deep-learning techniques for salient and category-specific object detection: a survey. IEEE Signal Processing Magazine, 2018, 35(1): 84–100

[7] Jiang H, Cheng M M, Li S J, Borji A, Wang J. Joint salient object detection and existence prediction. Frontiers of Computer Science, 2018, https://doi.org/10.1007/s11704-017-6613-8

[8] Girshick R B, Donahue J, Darrell T, Malik J. Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 2014, 580–587

[9] Ren S, He K, Girshick R, Sun J. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137–1149

[10] Liu W, Anguelov D, Erhan D, Szegedy C, Reed S E, Fu C, Berg A C. SSD: single shot multibox detector. In: Proceedings of European Conference on Computer Vision. Berlin: Springer, 2016, 21–37

[11] Redmon J, Divvala S K, Girshick R B, Farhadi A. You only look once: unified, real-time object detection. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016, 779–788

[12] Lin T Y, Dollar P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017, 936–944

[13] He K, Gkioxari G, Dollar P, Girshick R. Mask R-CNN. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2018, doi:10.1109/TPAMI.2018.2844175

[14] Everingham M, Eslami S M A, Gool L V, Williams C K I, Winn J, Zisserman A. The pascal visual object classes challenge: A Retrospective. International Journal of Computer Vision, 2015, 111(1): 98–136

[15] Uijlings J R R, van de Sande K E A, Gevers T, Smeulders A W M. Selective search for object recognition. International Journal of Computer Vision, 2013, 104(2): 154–171

[16] Felzenszwalb P F, Girshick R B, McAllester D, Ramanan D. Object detection with discriminatively trained part-based models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(9): 1627–1645

[17] Park H M, Cho D Y, Yoon K J. Greedy refinement of object proposals via boundary-aligned minimum bounding box search. IET Computer Vision, 2018, 12(3): 357–363

[18] Everingham M, Van Gool L, Williams C K I, Winn J, Zisserman A. The pascal visual object classes (VOC) challenge. International Journal of Computer Vision, 2010, 88(2): 303–338

[19] Geiger A, Lenz P, Urtasun R. Are we ready for autonomous driving? The KITTI vision benchmark suite. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Providence: IEEE, 2012, 3354–3361

[20] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. 2014, arXiv:1409.1556

[21] Chen Z, Zhang T, Ouyang C. End-to-end airplane detection using transfer learning in remote sensing images. Remote Sensing, 2018, 10(1): 139

[22] Pan S J, Yang Q. A survey on transfer learning. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10): 1345–1359

[23] Deng J, Dong W, Socher R, Li L J, Li K, Li F F. ImageNet: a large-scale hierarchical image database. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Miami: IEEE, 2009, 248–255

RIGHTS & PERMISSIONS

2019 Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature