Trilateral convolutional neural network for 3D shape reconstruction of objects from a single depth view

Rivera, Patricio; Añazco, Edwin Valarezo; Choi, Mun-Taek; Kim, Tae‐Seong

doi:10.1049/iet-ipr.2019.0532

Cited by 7 publications

(7 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meanwhile, the discriminator was dedicated to classifying real data and fake data. Based on the targets of the G and D modules, the objective function was set as Equation (3). The optimization criteria are to minimize the loss in training the generator and maximize the loss in training the discriminator.…”

Section: Training Detailsmentioning

confidence: 99%

“…Nowadays, 3D representation of real-world objects has become a core concept for vision, robotics, augmented reality and virtual reality applications, for its powerful geometric representation ability [1]. In computer vision and graphics, 3D object modelling can take on various forms of representation [2][3][4]. Hitherto, most researchers choose the 3D voxel grids or a collection of images to represent 3D object.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Constructing an efficient and adaptive learning model for 3D object generation

Deng

Liu

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

Studying representation learning and generative modelling has been at the core of the 3D learning domain. By leveraging the generative adversarial networks and convolutional neural networks for point-cloud representations, we propose a novel framework, which can directly generate 3D objects represented by point clouds. The novelties of the proposed method are threefold. First, the generative adversarial networks are applied to 3D object generation in the point-cloud space, where the model learns object representation from point clouds independently. In this work, we propose a 3D spatial transformer network, and integrate it into a generation model, whose ability for extracting and reconstructing features for 3D objects can be improved. Second, a point-wise approach is developed to reduce the computational complexity of the proposed network. Third, an evaluation system is proposed to measure the performance of our model by employing various categories and methods, and the error, considered as the difference between synthesized objects and raw objects are quantitatively compared, is less than 2.8%. Extensive experiments on benchmark dataset show that this method has a strong ability to generate 3D objects in the point-cloud space, and the synthesized objects have slight differences with man-made 3D objects. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

show abstract

Section: Training Detailsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Constructing an efficient and adaptive learning model for 3D object generation

Deng

Liu

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…The distance of each slide is called the convolution step [15]. In the process of convolution operation, for the input image and convolution kernel with a certain size, the number of feature extraction is determined by the step size, and the smaller the step size, the more times the feature is extracted [16].…”

Section: Cnn Structurementioning

confidence: 99%

Convolutional Neural Network in Image Recognition

2021

View full text Add to dashboard Cite

Image recognition is an important direction in computer vision. After years of research, image recognition technology has made great progress. This paper mainly studies the application of convolutional neural network (CNN) in image recognition. This paper first analyzes the basic mechanism of CNN, and on this basis, uses genetic algorithm to optimize CNN for image recognition application. Experimental results show that the optimal connection structure searched by this method can significantly improve the effect of CNN image recognition and speed up the training speed of the model.

show abstract

“…Additionally it was shown that by adopting Bernstein Basis Function Networks (BBFNs) it is also possible to solve the task of reconstructing a 3D shape [27]. A trilateral convolutional neural network (Tri-CNN) that uses three dilated convolutions in 3D to extend the convolutional receptive field was applied on the ShapeNet and Big Data for Grasp Planning [28] data sets to obtain 3D reconstruction from a single depth image [29].…”

Section: Introductionmentioning

confidence: 99%

3D Object Reconstruction from Imperfect Depth Data Using Extended YOLOv3 Network

Kulikajevas

Maskeliūnas

Damaševičius

et al. 2020

Sensors

View full text Add to dashboard Cite

State-of-the-art intelligent versatile applications provoke the usage of full 3D, depth-based streams, especially in the scenarios of intelligent remote control and communications, where virtual and augmented reality will soon become outdated and are forecasted to be replaced by point cloud streams providing explorable 3D environments of communication and industrial data. One of the most novel approaches employed in modern object reconstruction methods is to use a priori knowledge of the objects that are being reconstructed. Our approach is different as we strive to reconstruct a 3D object within much more difficult scenarios of limited data availability. Data stream is often limited by insufficient depth camera coverage and, as a result, the objects are occluded and data is lost. Our proposed hybrid artificial neural network modifications have improved the reconstruction results by 8.53% which allows us for much more precise filling of occluded object sides and reduction of noise during the process. Furthermore, the addition of object segmentation masks and the individual object instance classification is a leap forward towards a general-purpose scene reconstruction as opposed to a single object reconstruction task due to the ability to mask out overlapping object instances and using only masked object area in the reconstruction process.

show abstract

Trilateral convolutional neural network for 3D shape reconstruction of objects from a single depth view

Cited by 7 publications

References 18 publications

Constructing an efficient and adaptive learning model for 3D object generation

Constructing an efficient and adaptive learning model for 3D object generation

Convolutional Neural Network in Image Recognition

3D Object Reconstruction from Imperfect Depth Data Using Extended YOLOv3 Network

Contact Info

Product

Resources

About