[3] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 521, 436-444(2015).

[5] [5] PANG S C, DEL COZ J J, YU Z Z, et al. Combining deep learning preference learning f object tracking[C]Proceedings of the 23rd International Conference on Neural Infmation Processing. Kyoto, Japan: Springer, 2016.

[6] [6] GIRSHICK R. Fast RCNN[J]. Computer Science, 2015.

[7] [7] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, realtime object detection[C]Proceedings of 2016 IEEE Conference on Computer Vision Pattern Recognition. Las Vegas, USA: IEEE, 2016.

[8] [8] REDMON J, FARHADI A. YOLOv3: an incremental improvement[J]. arXiv EPrints, 2018, 108(4):625633.

[9] [9] ZHU X Z, HU H, LIN S, et al. Defmable Convs V2: me defmable, better results[C]Proceedings of 2019 IEEECVF Conference on Computer Vision Pattern Recognition. Long Beach, USA: IEEE, 2019.

[10] [10] DAI J F, QI H Z, XIONG Y W, et al. Defmable convolutional wks[J]. Computer Vision Pattern Recognition, 2017, 9(1):334420.

[11] [11] SADEGHI MA, FSYTH D. 30 Hz object detection with DPM V5[C]Proceedings of 13th European Conference on Computer Vision. Zurich, Switzerl: Springer, 2014.

[12] [12] REN S Q, HE K M, GIRSHICK R, et al. Faster RCNN: towards realtime object detection with region proposal wks[J].Neural Infmation Processing Systems, 2015, 452(1):108133.