Single MLP Research Articles

NU-NeRF: Neural Reconstruction of Nested Transparent Objects with Uncontrolled Capture Environment

The geometry reconstruction of transparent objects is a challenging problem due to the highly noncontinuous and rapidly changing surface color caused by refraction. Existing methods rely on special capture devices, dedicated backgrounds, or ground-truth object masks to provide more priors and reduce the ambiguity of the problem. However, it is hard to apply methods with these special requirements to real-life reconstruction tasks, like scenes captured in the wild using mobile devices. Moreover, these methods can only cope with solid and homogeneous materials, greatly limiting the scope of the application. To solve the problems above, we propose NU-NeRF to reconstruct nested transparent objects without requiring a dedicated capture environment or additional input. NU-NeRF is built upon a neural signed distance field formulation and leverages neural rendering techniques. It consists of two main stages. In Stage I, the surface color is separated into reflection and refraction. The reflection is decomposed using physically based material and rendering. The refraction is modeled using a single MLP given the refraction and view directions, which is a simple yet effective solution of refraction modeling. This step produces high-fidelity geometry of the outer surface. In stage II, we use explicit ray tracing on the reconstructed outer surface for accurate light transport simulation. The surface reconstruction is executed again inside the outer geometry to obtain any inner surface geometry. In this process, a novel transparent interface formulation is used to cope with different types of transparent surfaces. Experiments conducted on synthetic scenes and real captured scenes show that NU-NeRF is capable of producing better reconstruction results than previous methods and achieves accurate nested surface reconstruction under an uncontrolled capture environment.

Scalable neural indoor scene rendering

We propose a scalable neural scene reconstruction and rendering method to support distributed training and interactive rendering of large indoor scenes. Our representation is based on tiles. Tile appearances are trained in parallel through a background sampling strategy that augments each tile with distant scene information via a proxy global mesh. Each tile has two low-capacity MLPs: one for view-independent appearance (diffuse color and shading) and one for view-dependent appearance (specular highlights, reflections). We leverage the phenomena that complex view-dependent scene reflections can be attributed to virtual lights underneath surfaces at the total ray distance to the source. This lets us handle sparse samplings of the input scene where reflection highlights do not always appear consistently in input images. We show interactive free-viewpoint rendering results from five scenes, one of which covers an area of more than 100 m 2 . Experimental results show that our method produces higher-quality renderings than a single large-capacity MLP and five recent neural proxy-geometry and voxel-based baseline methods. Our code and data are available at project webpage https://xchaowu.github.io/papers/scalable-nisr.

NeuRec: Incorporating Interpatient prior to Sparse-View Image Reconstruction for Neurorehabilitation.

Medical imaging technologies such as computed tomography (CT) and magnetic resonance imaging (MRI) imaging are indispensable for contemporary neurorehabilitation diagnostics, intervention, and monitoring. It would be desirable to reconstruct images from sparse measurements to reduce the ionizing radiation and motion artifacts. Although recent coordinate-based representation methods have shown promise advances for sparse-view reconstruction, they overfit a single MLP on a single patient. In this work, we generalize it across many patients by incorporating an interpatient prior into the ill-posed inverse/reconstruction problem, which is the missing ingredient in the previous works. The experiment demonstrates that our method significantly improves image quality over the state-of-the-art both qualitatively and quantitatively. Thus, our method provides a powerful and principled means to deal with the measurement-scarce problem.

Pemanfaatan Big Data dalam Monitoring Pola Aktivitas Aviasi di Indonesia

Abstract: Covid-19 which entered Indonesia in December 2019 has a significant impact on the aviation industry. According to BPS data for 2020, the aviation industry's contribution to Indonesia's GDP decreased from 1.21% to 0.28% in the second quarter of 2020. To overcome this setback, comprehensive monitoring by policy makers is needed. The use of big data in monitoring aviation industry activities can be an option. This study aims to analyze aviation activities using big data approach for monitoring basis. The data was collected by using web scraping method on one of the global aviation websites to obtain flight status data at 108 airports in Indonesia on April 2020 until June 2021. Other data used are google mobility index data, GDP data, and TPK. The analysis method used are descriptive analysis, correlation analysis and machine learning based time series modelling with ARNN, single layer ANN and MLP. The results show that the policy of restricting mobility has a significant effect on the productivity of aviation industry. Machine learning modeling shows that the MLP model is the best model for forecasting international aviation activity. In addition, it was found that the aviation industry has a strong correlation with the economy and tourism sector in Indonesia.

Combining Neural Networks for Skin Detection

Two types of combining strategies were evaluated namely combining skin features and combining skin classifiers. Several combining rules were applied where the outputs of the skin classifiers are combined using binary operators such as the AND and the OR operators, Voting, Sum of Weights and a new neural network. Three chrominance components from the YCbCr colour space that gave the highest correct detection on their single feature MLP were selected as the combining parameters. A major issue in designing a MLP neural network is to determine the optimal number of hidden units given a set of training patterns. Therefore, a coarse to fine search method to find the number of neurons in the hidden layer is proposed. The strategy of combining Cb/Cr and Cr features improved the correct detection by 3.01% compared to the best single feature MLP given by Cb-Cr. The strategy of combining the outputs of three skin classifiers using the Sum of Weights rule further improved the correct detection by 4.38% compared to the best single feature MLP.

An Optimal SVM with Feature Selection Using Multiobjective PSO

Support vector machine is a classifier, based on the structured risk minimization principle. The performance of the SVM depends on different parameters such as penalty factor, C, and the kernel factor, σ. Also choosing an appropriate kernel function can improve the recognition score and lower the amount of computation. Furthermore, selecting the useful features among several features in dataset not only increases the performance of the SVM, but also reduces the computational time and complexity. So this is an optimization problem which can be solved by heuristic algorithm. In some cases besides the recognition score, the reliability of the classifier’s output is important. So in such cases a multiobjective optimization algorithm is needed. In this paper we have got the MOPSO algorithm to optimize the parameters of the SVM, choose appropriate kernel function, and select the best feature subset simultaneously in order to optimize the recognition score and the reliability of the SVM concurrently. Nine different datasets, from UCI machine learning repository, are used to evaluate the power and the effectiveness of the proposed method (MOPSO-SVM). The results of the proposed method are compared to those which are achieved by single SVM, RBF, and MLP neural networks.

Unsupervised machine-learning method for improving the performance of ambulatory fall-detection systems

BackgroundFalls can cause trauma, disability and death among older people. Ambulatory accelerometer devices are currently capable of detecting falls in a controlled environment. However, research suggests that most current approaches can tend to have insufficient sensitivity and specificity in non-laboratory environments, in part because impacts can be experienced as part of ordinary daily living activities.MethodWe used a waist-worn wireless tri-axial accelerometer combined with digital signal processing, clustering and neural network classifiers. The method includes the application of Discrete Wavelet Transform, Regrouping Particle Swarm Optimization, Gaussian Distribution of Clustered Knowledge and an ensemble of classifiers including a multilayer perceptron and Augmented Radial Basis Function (ARBF) neural networks.ResultsPreliminary testing with 8 healthy individuals in a home environment yields 98.6% sensitivity to falls and 99.6% specificity for routine Activities of Daily Living (ADL) data. Single ARB and MLP classifiers were compared with a combined classifier. The combined classifier offers the greatest sensitivity, with a slight reduction in specificity for routine ADL and an increased specificity for exercise activities. In preliminary tests, the approach achieves 100% sensitivity on in-group falls, 97.65% on out-group falls, 99.33% specificity on routine ADL, and 96.59% specificity on exercise ADL.ConclusionThe pre-processing and feature-extraction steps appear to simplify the signal while successfully extracting the essential features that are required to characterize a fall. The results suggest this combination of classifiers can perform better than MLP alone. Preliminary testing suggests these methods may be useful for researchers who are attempting to improve the performance of ambulatory fall-detection systems.

COMBINING NEURAL NETWORKS FOR SKIN DETECTION

Two types of combining strategies were evaluated namely combining skin features and combining skin classifiers. Several combining rules were applied where the outputs of the skin classifiers are combined using binary operators such as the AND and the OR operators, “Voting�, “Sum of Weights� and a new neural network. Three chrominance components from the YCbCr colour space that gave the highest correct detection on their single feature MLP were selected as the combining parameters. A major issue in designing a MLP neural network is to determine the optimal number of hidden units given a set of training patterns. Therefore, a “coarse to fine search� method to find the number of neurons in the hidden layer is proposed. The strategy of combining Cb /Cr and Cr features improved the correct detection by 3.01% compared to the best single feature MLP given by Cb-Cr. The strategy of combining the outputs of three skin classifiers using the “Sum of Weights� rule further improved the correct detection by 4.38% compared to the best single feature MLP.

MLP ensembles improve long term prediction accuracy over single networks

Abstract This work describes an award winning approach for solving the NN3 Forecasting Competition problem, focusing on the sound experimental validation of its main innovative feature. The NN3 forecasting task consisted of predicting 18 future values of 111 short monthly time series. The main feature of the approach was the use of the median for combining the forecasts of an ensemble of 15 MLPs to predict each time series. Experimental comparison to a single MLP shows that the ensemble increases the performance accuracy for multiple-step ahead forecasting. This system performed well on the withheld data, having finished as the second best solution of the competition with an SMAPE of 16.17%.

A fast algorithm for inversion of MLP networks in design problems

PurposeTo present a neural network‐based approach to the design of electromagnetic devices.Design/methodology/approachA neural model is created which reproduces the relationship between the design parameters of the device and the performance parameters, typically field values.FindingsThe neural model is a single hidden layer MLP network, trained by using a set of cases calculated, for example, by means of a finite element analysis. The design problem can be solved by fixing the performance values at the output of the network and by calculating the corresponding input values. The relationship between the input and the output of the neural network is represented by three equations systems. By means of these three systems, we can forward the domain of the input, and we can back propagate the desired output throughout the network layers. In such a way, both the domain of the design parameters and the domain of the desired performances values can be projected in the same space. Whatever point inside the intersection between the two projected domains corresponds to a solution of the design problem.Originality/valuePresents a procedure which is able to find a point belonging to such an intersection.

An experimental analysis of the dependence among codeword bit errors in ECOC learning machines

One of the main factors affecting the effectiveness of error correcting output coding (ECOC) methods for classification is the dependence among the errors of the computed codeword bits. We present an extensive experimental work for evaluating the dependence among output errors of the decomposition unit in ECOC learning machines. In particular, we apply measures based on mutual information to compare the dependence of ECOC multi-layer perceptron (ECOC MLP), made up by a single multi-input multi-output MLP, and ECOC ensembles made up by a set of independent and parallel dichotomizers (ECOC PND). Moreover, the experimentation analyzes the relationship between the architecture, the dependence among output errors and the performances of ECOC learning machines. The results show that the dependence among computed codeword bits is significantly smaller for ECOC PND, pointing out that ensembles of independent parallel dichotomizers are better suited for implementing ECOC classification methods. The experimental results suggest new architectures of ECOC learning machines to improve the independence among output errors and the diversity between base learners.

Lower bounds for approximation by MLP neural networks

The degree of approximation by a single hidden layer MLP model with n units in the hidden layer is bounded below by the degree of approximation by a linear combination of n ridge functions. We prove that there exists an analytic, strictly monotone, sigmoidal activation function for which this lower bound is essentially attained. We also prove, using this same activation function, that one can approximate arbitrarily well any continuous function on any compact domain by a two hidden layer MLP using a fixed finite number of units in each layer.