Abstract
As a representative technology of artificial intelligence, 3D reconstruction based on deep learning can be integrated into the edge computing framework to form an intelligent edge and then realize the intelligent processing of the edge. Recently, high-resolution representation of 3D objects using multiview decomposition (MVD) architecture is a fast reconstruction method for generating objects with realistic details from a single RGB image. The results of high-resolution 3D object reconstruction are related to two aspects. On the one hand, a low-resolution reconstruction network represents a good 3D object from a single RGB image. On the other hand, a high-resolution reconstruction network maximizes fine low-resolution 3D objects. To improve these two aspects and further enhance the high-resolution reconstruction capabilities of the 3D object generation network, we study and improve the low-resolution 3D generation network and the depth map superresolution network. Eventually, we get an improved multiview decomposition (IMVD) network. First, we use a 2D image encoder with multifeature fusion (MFF) to enhance the feature extraction capability of the model. Second, a 3D decoder using an effective subpixel convolutional neural network (3D ESPCN) improves the decoding speed in the decoding stage. Moreover, we design a multiresidual dense block (MRDB) to optimize the depth map superresolution network, which allows the model to capture more object details and reduce the model parameters by approximately 25% when the number of network layers is doubled. The experimental results show that the proposed IMVD is better than the original MVD in the 3D object superresolution experiment and the high-resolution 3D reconstruction experiment of a single image.
Highlights
The three-dimensional reconstruction of a single image is a hotspot and a difficult point in the field of computer vision
Using 3D ESPCN can generate 3D shapes in lower resolution 3D volume spaces than traditional 3D decoders in the last step of the 3D decoding stage. This reduces the time required for the model to generate 3D shapes (iii) We propose a multiresidual dense network to make full use of the features extracted from the residual network and the dense network
We show the experimental results of the improved multiview decomposition (IMVD) network for 3D object superresolution and 3D object reconstruction of a single RGB image
Summary
The three-dimensional reconstruction of a single image is a hotspot and a difficult point in the field of computer vision. In addition to the direct use of voxel methods to generate 3D shapes, other studies have used different three-dimensional representations, such as point clouds [28,29,30], meshes [31,32,33], primitives [34, 35], and implicit surfaces [36, 37]. Most of these methods can reconstruct three-dimensional objects with high resolution and are not limited by memory requirements. Most of these methods need to solve the inherent defects of the model, such as using the point cloud method to reconstruct the surface details of the object and solving the genus problem of the mesh method to reconstruct the object
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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.
