RipViz: Finding Rip Currents by Learning Pathline Behavior.

We present a hybrid machine learning and flow analysis feature detection method, RipViz, to extract rip currents from stationary videos. Rip currents are dangerous strong currents that can drag beachgoers out to sea. Most people are either unaware of them or do not know what they look like. In some instances, even trained personnel such as lifeguards have difficulty identifying them. RipViz produces a simple, easy to understand visualization of rip location overlaid on the source video. With RipViz, we first obtain an unsteady 2D vector field from the stationary video using optical flow. Movement at each pixel is analyzed over time. At each seed point, sequences of short pathlines, rather a single long pathline, are traced across the frames of the video to better capture the quasi-periodic flow behavior of wave activity. Because of the motion on the beach, the surf zone, and the surrounding areas, these pathlines may still appear very cluttered and incomprehensible. Furthermore, lay audiences are not familiar with pathlines and may not know how to interpret them. To address this, we treat rip currents as a flow anomaly in an otherwise normal flow. To learn about the normal flow behavior, we train an LSTM autoencoder with pathline sequences from normal ocean, foreground, and background movements. During test time, we use the trained LSTM autoencoder to detect anomalous pathlines (i.e., those in the rip zone). The origination points of such anomalous pathlines, over the course of the video, are then presented as points within the rip zone. RipViz is fully automated and does not require user input. Feedback from domain expert suggests that RipViz has the potential for wider use.

Stable Feature Flow Fields.

Feature Flow Fields are a well-accepted approach for extracting and tracking features. In particular, they are often used to track critical points in time-dependent vector fields and to extract and track vortex core lines. The general idea is to extract the feature or its temporal evolution using a stream line integration in a derived vector field-the so-called Feature Flow Field (FFF). Hence, the desired feature line is a stream line of the FFF. As we will carefully analyze in this paper, the stream lines around this feature line may diverge from it. This creates an unstable situation: if the integration moves slightly off the feature line due to numerical errors, then it will be captured by the diverging neighborhood and carried away from the real feature line. The goal of this paper is to define a new FFF with the guarantee that the neighborhood of a feature line has always converging behavior. This way, we have an automatic correction of numerical errors: if the integration moves slightly off the feature line, it automatically moves back to it during the ongoing integration. This yields results which are an order of magnitude more accurate than the results from previous schemes. We present new stable FFF formulations for the main applications of tracking critical points and solving the Parallel Vectors operator. We apply our method to a number of data sets.

Using PVsolve to analyze and locate positions of parallel vectors.

A new method for finding the locus of parallel vectors is presented, called PVsolve. A parallel-vector operator has been proposed as a visualization primitive, as several features can be expressed as the locus of points where two vector fields are parallel. Several applications of the idea have been reported, so accurate and efficient location of such points is an important problem. Previously published methods derive a tangent direction under the assumption that the two vector fields are parallel at the current point in space, then extend in that direction to a new point. PVsolve includes additional terms to allow for the fact that the two vector fields may not be parallel at the current point, and uses a root-finding approach. Mathematical analysis sheds new light on the feature flow field technique (FFF) as well. The root-finding property allows PVsolve to use larger step sizes for tracing parallel-vector curves, compared to previous methods, and does not rely on sophisticated differential equation techniques for accuracy. Experiments are reported on fluid flow simulations, comparing FFF and PVsolve.

Topological lines in 3D tensor fields and discriminant Hessian factorization.

This paper addresses several issues related to topological analysis of 3D second order symmetric tensor fields. First, we show that the degenerate features in such data sets form stable topological lines rather than points, as previously thought. Second, the paper presents two different methods for extracting these features by identifying the individual points on these lines and connecting them. Third, this paper proposes an analytical form of obtaining tangents at the degenerate points along these topological lines. The tangents are derived from a Hessian factorization technique on the tensor discriminant and leads to a fast and stable solution. Together, these three advances allow us to extract the backbone topological lines that form the basis for topological analysis of tensor fields.

Comparative flow visualization.

There are many situations where one needs to compare two or more data sets. It may be to compare different models, different resolutions, differences in algorithms, different experimental results, etc. There is therefore a need for comparative visualization tools to help analyze the differences. This paper focuses on comparative visualization tools for analyzing flow or vector data sets. The techniques presented allow one to compare individual streamlines and streamribbons as well as a dense field of streamlines. These comparison methods can also be used to study differences in vortex cores that are represented as polylines.