Perception techniques for autonomous driving should be adaptive to various environments. In essential perception modules for traffic line detection, many conditions should be considered, such as a number of traffic lines and computing power of the target system. To address these problems, in this paper, we propose a traffic line detection method called Point Instance Network (PINet); the method is based on the key points estimation and instance segmentation approach. The PINet includes several hourglass models that are trained simultaneously with the same loss function. Therefore, the size of the trained models can be chosen according to the target environment’s computing power. We cast a clustering problem of the predicted key points as an instance segmentation problem; the PINet can be trained regardless of the number of the traffic lines. The PINet achieves competitive accuracy and false positive on CULane and TuSimple datasets, popular public datasets for lane detection. Our code is available at https://github.com/koyeongmin/PINet_new

In today's vector space information retrieval systems, dimension reduction is imperative for efficiently manipulating the massive quantity of data. To be useful, this lower-dimensional representation must be a good approximation of the full document set. To that end, we adapt and extend the discriminant analysis projection used in pattern recognition. This projection preserves cluster structure by maximizing the scatter between clusters while minimizing the scatter within clusters. A common limitation of trace optimization in discriminant analysis is that one of the scatter matrices must be nonsingular, which restricts its application to document sets in which the number of terms does not exceed the number of documents. We show that by using the generalized singular value decomposition (GSVD), we can achieve the same goal regardless of the relative dimensions of the term-document matrix. In addition, applying the GSVD allows us to avoid the explicit formation of the scatter matrices in favor of working di...

Background: Atrial fibrillation (AFib) is one of the prominent causes of stroke, and its risk increases with age. We need to detect AFib correctly as early as possible to avoid medical disaster because it is likely to proceed into a more serious form in short time. If we can make a portable AFib monitoring system, it will be helpful to many old people because we cannot predict when a patient will have a spasm of AFib. Methods: We analyzed heart beat variability from inter-beat intervals obtained by a wavelet-based detector. We made a Poincare plot using the inter-beat intervals. By analyzing the plot, we extracted three feature measures characterizing AFib and non-AFib: the number of clusters, mean stepping increment of inter-beat intervals, and dispersion of the points around a diagonal line in the plot. We divided distribution of the number of clusters into two and calculated mean value of the lower part by k-means clustering method. We classified data whose number of clusters is more than one and less than this mean value as non-AFib data. In the other case, we tried to discriminate AFib from non-AFib using support vector machine with the other feature measures: the mean stepping increment and dispersion of the points in the Poincare plot. Results: We found that Poincare plot from non-AFib data showed some pattern, while the plot from AFib data showed irregularly irregular shape. In case of non-AFib data, the definite pattern in the plot manifested itself with some limited number of clusters or closely packed one cluster. In case of AFib data, the number of clusters in the plot was one or too many. We evaluated the accuracy using leave-one-out cross-validation. Mean sensitivity and mean specificity were 91.4% and 92.9% respectively. Conclusions: Because pulse beats of ventricles are less likely to be influenced by baseline wandering and noise, we used the inter-beat intervals to diagnose AFib. We visually displayed regularity of the inter-beat intervals by way of Poincare plot. We tried to design an automated algorithm which did not require any human intervention and any specific threshold, and could be installed in a portable AFib monitoring system.

We present adversarial event prediction (AEP), a novel approach to detecting abnormal events through an event prediction setting. Given normal event samples, AEP derives the prediction model, which can discover the correlation between the present and future of events in the training step. In obtaining the prediction model, we propose adversarial learning for the past and future of events. The proposed adversarial learning enforces AEP to learn the representation for predicting future events and restricts the representation learning for the past of events. By exploiting the proposed adversarial learning, AEP can produce the discriminative model to detect an anomaly of events without complementary information, such as optical flow and explicit abnormal event samples in the training step. We demonstrate the efficiency of AEP for detecting anomalies of events using the UCSD-Ped, CUHK Avenue, Subway, and UCF-Crime data sets. Experiments include the performance analysis depending on hyperparameter settings and the comparison with existing state-of-the-art methods. The experimental results show that the proposed adversarial learning can assist in deriving a better model for normal events on AEP, and AEP trained by the proposed adversarial learning can surpass the existing state-of-the-art methods.