Abstract

Object detection algorithm based on depth model has achieved state-of-the-art results on various challenging benchmarks. However, the large amount of parameters of depth model means a large amount of calculation. This seriously limits the practical application of object detection algorithm, especially on embedded devices with limited computing power. We propose an object detection algorithm based on DenseNet and Region Proposal Network(RPN) and replace ROI Pooling with ROI Align. From the evaluation on PASCAL VOC and MS COCO we can see that the algorithm achieves object detection with fewer parameters while maintaining or improving accuracy. This is meaningful for the development of embedded in-depth learning. Finally, we explore the influence of different shared convolutional layers on object detection algorithm.

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.