Abstract
In this paper, the object recognition module of a visual auxiliary system, InVision, called IR-VP, is presented. (1) Deep multimodal neural network (DMNN) is presented to enhance CNNs’ sampling abilitiy and feature resolutions. (2) Deep distance learning (DDL) is presented to find similarity and reduce representation redundancy. (3) The light weight MobileNet is used to accelerate. Extensive experiments demonstrate that the proposed approach significantly outperforms state-of-the-arts.
Highlights
Vision plays an important role in every one’s life.Reduced eyesight has serious effects on all aspects of life, if we cannot see, it is difficult for us to learn to walk, finish daily personal activities, interact with the community, school and work opportunities and the ability to access public services.The fact is that eye conditions are remarkably common
We develop a novel visual auxiliary system, named InVision, based on bone conduction, speech, radar technology, ergonomic design, and advanced deep learning algorithm
We mainly introduce the object recognition module of InVision — IR-VP, based on radar and deep learning and humanized-design
Summary
Vision plays an important role in every one’s life. Reduced eyesight has serious effects on all aspects of life, if we cannot see, it is difficult for us to learn to walk, finish daily personal activities, interact with the community, school and work opportunities and the ability to access public services. (1) Deep multimodal neural network (DMNN) is presented to enhance the CNNs’ perception, sampling ability and feature resolutions. (2) Deep distance learning (DDL) is presented find similarity and reduce representation redundancy. A. Overall Framework There are three main modules in IR-VP: (1) deep multimodal neural network (DMNN), (2) deep distance learning (DDL), and (3) MobileNet. FIGURE 1. B. Deep Multimodal Convolutional Neural Network Dilated deformable convolution (DDC) is proposed in DMNN to enlarge the receptive field sizes of kernels and output feature maps’ resolutions. Regions ranking and matching module is proposed to reduce the redundancy and improve the efficiency. To reduce the redundancy, improve the efficiency, regions ranking and matching modules [13,14,15,16] are presented, which consider the appearance and geometric relationship to achieve reliable correspondences
Sign-in/Register to view highlights & summary 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 callDisclaimer: 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.
