Presentation is loading. Please wait.

Presentation is loading. Please wait.

Advanced Light and Shadow Culling Methods Eric Lengyel.

Similar presentations


Presentation on theme: "Advanced Light and Shadow Culling Methods Eric Lengyel."— Presentation transcript:

1

2 Advanced Light and Shadow Culling Methods Eric Lengyel

3 Fully Dynamic Environment Anything in the world can move Anything in the world can move — Can’t precompute any visibility information Lights completely dynamic Lights completely dynamic — Can’t precompute any lighting information — Shadows also completely dynamic

4 Problems to Be Solved at Run-time Determine the set of objects visible to the camera Determine the set of objects visible to the camera Determine the set of lights that can influence the region of space visible to the camera Determine the set of lights that can influence the region of space visible to the camera For each light, also determine what subset of the visible objects can be illuminated For each light, also determine what subset of the visible objects can be illuminated

5 Problems to Be Solved at Run-time Determine the set of objects that could possibly cast shadows into the region of space visible to the camera Determine the set of objects that could possibly cast shadows into the region of space visible to the camera — A superset of the set of the illuminated objects that are visible to the camera

6 Visibility Determination Organize the world in some way Organize the world in some way — Tree structures (BSP, octree, etc.) — Hierarchical bounding volumes — Portal system A combination of these can work extremely well A combination of these can work extremely well

7 Portal Systems World divided into ‘zones’ World divided into ‘zones’ — A zone is the region of space bounded by a convex polyhedron Zones are connected by ‘portals’ Zones are connected by ‘portals’ — A portal is a planar convex polygon — From the front side, a portal’s vertices are wound CCW

8 Portal Systems During visibility determination, only have to worry about zones that can be seen through a sequence of portals During visibility determination, only have to worry about zones that can be seen through a sequence of portals For each reachable zone, there is a convex region of space visible to the camera For each reachable zone, there is a convex region of space visible to the camera

9 Portal Systems Camera

10 The visible regions form a tree structure The visible regions form a tree structure The region in the zone containing the camera is the root of the tree The region in the zone containing the camera is the root of the tree Zones seen through n portals have regions at the n-th level in the tree Zones seen through n portals have regions at the n-th level in the tree

11 Portal Systems Camera A B C D A BC D

12 Start with view frustum in zone containing the camera Start with view frustum in zone containing the camera Clip portals leading out of the zone against the view frustum Clip portals leading out of the zone against the view frustum For any portals intersecting the view frustum, construct a new region by extruding clipped edges away from camera position For any portals intersecting the view frustum, construct a new region by extruding clipped edges away from camera position

13 Portal Systems Every visibility region is bounded by a convex polyhedron defined by at least 4 planes Every visibility region is bounded by a convex polyhedron defined by at least 4 planes — At least 3 lateral planes — A back plane — A front plane in non-root nodes Plane normals point inward Plane normals point inward

14 Portal Systems After region tree has been built... After region tree has been built... — Traverse the tree — Collect objects in each zone that intersect the visible region of space for that zone  Use any frustum/bounding volume test, but test against region’s planes — This is the visible object set

15 Light Region Trees Portals can be used to construct illumination trees Portals can be used to construct illumination trees — Similar to the visibility tree constructed for the camera — One tree for each light source  Only recalculated when light moves — Each node in the tree corresponds to a convex region of space

16 Light Region Trees Three fundamental light types Three fundamental light types — Point light — Spot light, special case of point light — Infinite (directional) light

17 Light Region Trees Point light Point light — Omnidirectional — Has maximum range — Root illumination region bounded only be zone boundary and light’s bounding sphere

18 Point Light Tree A B C D B AD C

19 Spot Light Tree Spot light almost same as point light Spot light almost same as point light — Difference is the root node of the illumination tree — Spot light starts with a frustum, just like a camera does — Point light affects entire root zone

20 Infinite Light Tree Light rays parallel for infinite light Light rays parallel for infinite light — The lateral planes of each illumination region intersect at parallel lines — The extrusion of planes from a portal always goes in one direction instead of away from a point

21 Visible Light Determination Each zone keeps a linked list of illumination regions Each zone keeps a linked list of illumination regions — One or more region nodes for each light that can shine into the zone — Each region knows which light generated it

22 Visible Light Determination AB CD Light 1 Light 2 Light 3 For example, consider zone C

23 Visible Light Determination B AD C A BC D D CB A Light 1 Light 2 Light 3

24 Visible Light Determination For any given zone, we can walk the linked list of illumination regions and collect unique lights For any given zone, we can walk the linked list of illumination regions and collect unique lights Repeat process for all zones referenced in the camera’s visibility tree Repeat process for all zones referenced in the camera’s visibility tree We now have the set of visible lights We now have the set of visible lights

25 Illuminated Object Set Given one visible zone and one visible light shining into that zone… Given one visible zone and one visible light shining into that zone… — Illuminated objects are those which intersect both a camera region and a light region

26 Illuminated Object Set Light Camera

27 Objects are often only partially within an illumination region Objects are often only partially within an illumination region — Lighting the whole object wastes rendering time due to extra fill — Fortunately, hardware provides an opportunity for optimization

28 Lighting Optimization Use hardware scissor rectangle Use hardware scissor rectangle — Calculate intersections of camera regions and illumination regions — Camera-space bounding box determines scissor rectangle GL_EXT_depth_bounds_test GL_EXT_depth_bounds_test — Works like a z axis for scissor box

29 Lighting Optimization Camera Image Plane Max Depth Min Depth

30 Lighting Optimization Scissor rectangle and depth bounds test Scissor rectangle and depth bounds test — Limits rendering for a single light to the maximal visible extents — Can also be applied to stencil shadow volumes

31 Shadow Map Optimization Same optimization can be applied from light’s perspective when rendering shadow maps Same optimization can be applied from light’s perspective when rendering shadow maps — Transform vertices of region intersections into light space — Calculate scissor rectangle and depth bounds — Reduces unnecessary depth fill

32 Shadow Map Optimization Light Image Plane Max Depth Min Depth

33 Shadow-Casting Object Set All objects in the illuminated set are also in the shadow-casting set All objects in the illuminated set are also in the shadow-casting set — But an object doesn’t have to be visible to be casting a shadow into one of the visible camera regions — The shadow-casting set is a superset of the illuminated set

34 Shadow-Casting Object Set Need to find objects between visible regions and light source Need to find objects between visible regions and light source We already have a structure in place to make this easy We already have a structure in place to make this easy From an illumination region, walk up the light’s tree to the root From an illumination region, walk up the light’s tree to the root

35 Shadow-Casting Object Set Camera Light ABC DE C EB A

36 Collect objects in branch of illumination tree containing visible regions Collect objects in branch of illumination tree containing visible regions — Can use lateral bounding planes of deepest visible regions

37 Shadow-Casting Object Set Camera Light CulledCaster

38 Where illumination regions and camera regions intersect... Where illumination regions and camera regions intersect... — Some objects inside illumination region, but not between camera region and light source

39 Shadow-Casting Object Set ExtraCaster

40 Eliminating shadow-casting objects on wrong side of camera region Eliminating shadow-casting objects on wrong side of camera region — Test objects against planes of camera region having a positive dot product with light position — For infinite light, use planes having a positive dot product with direction to light

41 Shadow-Casting Object Set

42 Demonstrations

43 Questions? lengyel@terathon.com lengyel@terathon.com Slides available at Slides available athttp://www.terathon.com/


Download ppt "Advanced Light and Shadow Culling Methods Eric Lengyel."

Similar presentations


Ads by Google