There are no reviews yet. Be the first to send feedback to the community and the maintainers!
visibility_polygon.js version 1.9 This code is released into the public domain - attribution is appreciated but not required. Made by Byron Knoll. https://github.com/byronknoll/visibility-polygon-js Demo: http://www.byronknoll.com/visibility.html This library can be used to construct a visibility polygon for a set of line segments. The time complexity of this implementation is O(N log N) (where N is the total number of line segments). This is the optimal time complexity for this problem. The following functions should be useful: 1) VisibilityPolygon.compute(position, segments) Computes a visibility polygon. O(N log N) time complexity (where N is the number of line segments). Arguments: position - The location of the observer. If the observer is not completely surrounded by line segments, an outer bounding-box will be automatically created (so that the visibility polygon does not extend to infinity). segments - A list of line segments. Each line segment should be a list of two points. Each point should be a list of two coordinates. Line segments can not intersect each other. Overlapping vertices are OK, but it is not OK if a vertex is touching the middle of a line segment. Use the "breakIntersections" function to fix intersecting line segments. Returns: The visibility polygon (in clockwise vertex order). 2) VisibilityPolygon.computeViewport(position, segments, viewportMinCorner, viewportMaxCorner) Computes a visibility polygon within the given viewport. This can be faster than the "compute" function if there are many segments outside of the viewport. Arguments: position - The location of the observer. Must be within the viewport. segments - A list of line segments. Line segments can not intersect each other. It is OK if line segments intersect the viewport. viewportMinCorner - The minimum X and Y coordinates of the viewport. viewportMaxCorner - The maximum X and Y coordinates of the viewport. Returns: The visibility polygon within the viewport (in clockwise vertex order). 3) VisibilityPolygon.inPolygon(position, polygon) Calculates whether a point is within a polygon. O(N) time complexity (where N is the number of points in the polygon). Arguments: position - The point to check: a list of two coordinates. polygon - The polygon to check: a list of points. The polygon can be specified in either clockwise or counterclockwise vertex order. Returns: True if "position" is within the polygon. 4) VisibilityPolygon.convertToSegments(polygons) Converts the given polygons to list of line segments. O(N) time complexity (where N is the number of polygons). Arguments: a list of polygons (in either clockwise or counterclockwise vertex order). Each polygon should be a list of points. Each point should be a list of two coordinates. Returns: a list of line segments. 5) VisibilityPolygon.breakIntersections(segments) Breaks apart line segments so that none of them intersect. O(N^2) time complexity (where N is the number of line segments). Arguments: a list of line segments. Each line segment should be a list of two points. Each point should be a list of two coordinates. Returns: a list of line segments. Example code: var polygons = []; polygons.push([[-1,-1],[501,-1],[501,501],[-1,501]]); polygons.push([[250,100],[260,140],[240,140]]); var segments = VisibilityPolygon.convertToSegments(polygons); segments = VisibilityPolygon.breakIntersections(segments); var position = [60, 60]; if (VisibilityPolygon.inPolygon(position, polygons[0])) { var visibility = VisibilityPolygon.compute(position, segments); } var viewportVisibility = VisibilityPolygon.computeViewport(position, segments, [50, 50], [450, 450]);
cmix
cmix is a lossless data compression program aimed at optimizing compression ratio at the cost of high CPU/memory usage.lstm-compress
Data compression using LSTMtensorflow-compress
Data compression using LSTM in TensorFlowbattle-map-explorer
Create maps for D&D and other tabletop RPGs. Improved map exploration by hiding undiscovered rooms until they're entered.space-saving
This is a C++ implementation of the "space-saving" algorithm.predcast
This project uses pattern recognition to perform time series forecasting.paqclass
PAQclass is a classification algorithm based on a powerful compression program called PAQ8.pytorch-compress
Data compression using LSTMLove Open Source and this site? Check out how you can help us