LightingGridHierarchy Class Reference

Detailed Description

An implementation of the Lighting Grid Hierarchy method.

Can Yuksel and Cem Yuksel. 2017. Lighting Grid Hierarchy for Self-illuminating Explosions. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2017) 36, 4, Article 110 (July 2017). http://www.cemyuksel.com/research/lgh/

This class builds the Lighting Grid Hierarchy and uses it for lighting.

#include <cyLightingGrid.h>

Public Member Functions
	LightingGridHierarchy ()
	Constructor.
virtual	~LightingGridHierarchy ()
	Destructor.
int	GetNumLevels () const
	Returns the number of levels in the hierarchy.
float	GetCellSize () const
	Returns the size of a cell in the lowest (finest) level of the hierarchy.
const Vec3f &	GetLightPos (int level, int i) const
	Returns the i^th light position at the given level. Note that this is not the position of the light with index i.
const int	GetLightIndex (int level, int i) const
	Returns the i^th light index at the given level.
const Color &	GetLightIntens (int level, int ix) const
	Returns the intensity of the light with index ix at the given level.
const Vec3f &	GetLightPosDev (int level, int ix) const
	Returns the position variation of the light with index ix at the given level.
void	Clear ()
	Deletes all data.
bool	Build (const Vec3f *lightPos, const Color *lightIntensities, int numLights, int minLevelLights, float cellSize, int highestLevel)
	Builds the Lighting Grid Hierarchy for the given point light positions and intensities using the given parameters. This method builds the hierarchy using the given cellSize as the size of the lowest (finest) level grid cells.
bool	Build (const Vec3f *lightPos, const Color *lightIntensities, int numLights, int minLevelLights, float autoFitScale=1.01f)
	Builds the Lighting Grid Hierarchy for the given point light positions and intensities using the given parameters. This method automatically determines the grid cell size based on the bounding box of the light positions.
template<typename LightingFunction >
void	Light (const Vec3f &pos, float alpha, LightingFunction lightingFunction)
	Computes the illumination at the given position using the given accuracy parameter alpha.
template<typename LightingFunction >
void	Light (const Vec3f &pos, float alpha, int stochasticShadowSamples, LightingFunction lightingFunction)
	Computes the illumination at the given position using the given accuracy parameter alpha. This method provides stochastic sampling by randomly changing the given light position when calling the lighting function.

Member Function Documentation

Build() [1/2]

bool Build	(	const Vec3f *	lightPos,
		const Color *	lightIntensities,
		int	numLights,
		int	minLevelLights,
		float	cellSize,
		int	highestLevel )

Builds the Lighting Grid Hierarchy for the given point light positions and intensities using the given parameters. This method builds the hierarchy using the given cellSize as the size of the lowest (finest) level grid cells.

Parameters

lightPos	Light positions.
lightIntensities	Light intensities.
numLights	Number of lights.
minLevelLights	The minimum number of lights permitted for the highest (coarsest) level of the hierarchy. The build stops if a higher (coarser) level would have fewer lights.
cellSize	The size of a grid cell in the lowest (finest) level of the hierarchy.
highestLevel	The highest level permitted, where level 0 contains the original lights.

Build() [2/2]

bool Build	(	const Vec3f *	lightPos,
		const Color *	lightIntensities,
		int	numLights,
		int	minLevelLights,
		float	autoFitScale = 1.01f )

Builds the Lighting Grid Hierarchy for the given point light positions and intensities using the given parameters. This method automatically determines the grid cell size based on the bounding box of the light positions.

Parameters

lightPos	Light positions.
lightIntensities	Light intensities.
numLights	Number of lights.
minLevelLights	The minimum number of lights permitted for the highest (coarsest) level of the hierarchy. The build stops if a higher (coarser) level would have fewer lights.
autoFitScale	Extends the bounding box of the light positions using the given scale. This value must be 1 or greater. A value slightly greater than 1 often provides a good fit for the grid.

Light() [1/2]

template<typename LightingFunction >

void Light	(	const Vec3f &	pos,
		float	alpha,
		LightingFunction	lightingFunction )

Computes the illumination at the given position using the given accuracy parameter alpha.

Parameters

pos	The position where the lighting will be evaluated.
alpha	The accuracy parameter. It should be 1 or greater. Larger values produce more accurate results with substantially more computation.
lightingFunction	This function is called for each light used for lighting computation. It should be in the form void LightingFunction(int level, int light_id, const Vec3f &light_position, const Color &light_intensity).

Light() [2/2]

template<typename LightingFunction >

void Light	(	const Vec3f &	pos,
		float	alpha,
		int	stochasticShadowSamples,
		LightingFunction	lightingFunction )

Computes the illumination at the given position using the given accuracy parameter alpha. This method provides stochastic sampling by randomly changing the given light position when calling the lighting function.

Parameters

pos	The position where the lighting will be evaluated.
alpha	The accuracy parameter. It should be 1 or greater. Larger values produce more accurate results with substantially more computation.
stochasticShadowSamples	When this parameter is zero, the given lightingFunction is called once per light, using the position of the light. Otherwise, it is called as many times as this parameter specifies, using random positions around each light position.
lightingFunction	This function is called for each light used for lighting computation. It should be in the form void LightingFunction(int level, int light_id, const Vec3f &light_position, const Color &light_intensity).