Figure 2 - uploaded by Peter M Hall
Content may be subject to copyright.
A schematic view of Computer Graphics, Computer Vision, and Machine Learning. The elements in blue are inputs to some computer program, the element in red is what the program produces.
Source publication
Computer Graphics is increasingly using techniques from Machine Learning. The trend is motivated by several factors, but the difficulties and expense of modelling is a major driving force. Here 'modelling' is used very broadly to include models of reflection (learn the BRDF of a real material), animation (learn the motion of real objects), as well...
Contexts in source publication
Context 1
... example shows how learned probabilistic rules can replace hardcoded rules of production. Machine Learning is somehow orthogonal to both, as depicted in Figure 2. ...
Context 2
... doing so there are two competing demands to be met. The first is to optimally fill the volume into which the tree must 'grow', the second is to prevent tree branches from being bent into implausible shapes, as in Figure 20. Li et al. balance these demands using Bayes' law, as in Equation 15. ...
Similar publications
A traffic flow time series is a sequence of traffic detection parameters in chronological order. This differs from a general quantitative data sequence in that the time series includes a time attribute that contains not only the data with time characteristics, but also the distribution of the data itself. To date, studies of traffic time series hav...
In developing nations, many expanding cities are facing challenges that result from the overwhelming numbers of people and vehicles. Collecting real-time, reliable and precise traffic flow information is crucial for urban traffic management. The main purpose of this paper is to develop an adaptive model that can assess the real-time vehicle counts...
A traffic tensor or simply is a new data model for conventional origin/destination (O/D) matrices. Tensor models are traffic data analysis techniques which use this new data model to improve performance. Tensors outperform other models because both temporal and spatial fluctuations of traffic patterns are simultaneously taken into account, obtainin...
This paper introduces a binary neural network-based prediction algorithm incorporating both spatial and temporal characteristics into the prediction process. The algorithm is used to predict short-term traffic flow by combining information from multiple traffic sensors (spatial lag) and time series prediction (temporal lag). It extends previously d...
An important function of traffic flow management is ensuring the number of aircraft entering a sector does not exceed the amount that can be safely controlled by the sector controller. One factor that makes this task difficult is the uncertainty of the impact of convective weather, as both the weather forecast and the impact given specific weather...
Citations
... The fulfillment of these objectives would qualify the mapping for numerous applications in various fields ranging from econometrics, biometrics, and technometrics to pattern recognition and computer graphics, 15 which increasingly focuses on real-time construction of models from real-world data. 16 A literature survey on previous work on parameter estimation revealed the only method 17 to overcome the lack of a computationally fast, robust, and reliable curve fitting, which takes nonpositive observations into account without any data transformations. It is addressed to the special class of exponential models with an additive constant. ...
... It should be noted that positive y 0 i always fulfill the range of (18), which is due to (16). ...
... If sets of scattered data points x i , y o i À Á are equally spaced by x i ¼ i ¼ 1, …, n, Faulhaber equations can be used to calculate the variance term in (23), and the mean-value form, of a straight-line segment of Definition 1 becomes using (16): ...
A centroid‐ and covariance‐invariant deterministic mapping of sets of discrete data points to nonlinear models is introduced. Conditions for bijectivity of this mapping are developed. Since it can be accomplished by look‐up tables for the special case of equally spaced data, the resulting mapping algorithm is considered computationally fast. This is attractive for real‐time parameter estimation without the need of iterations and initial guesses of parameter values. Examples show that model parameter identification is easier to apply than by nonlinear least squares regression. Further, the approach is superior to log‐linear regression since it may allow to handle nonpositive observations without any transformations.
In this investigation we discuss a multi-label classification problem where documents may have several labels. We put our focus on dependencies among labels in a probabilistic manner, and we extract characteristic features in a form of probabilistic distribution functions by data mining techniques. We show some experimental results, i.e., dependencies among items/labels to see the effectiveness of the approach.
