OpenSubdiv/opensubdiv/sdc/crease.h
manuelk ce55f56d50 Rename Sdc::Options vertex boundary interpolation rule enums
- VVarBoundaryInterpolation is now VtxBoundaryInterpolation
- enum prefix change from VVAR to VTX
- generel cleanup / doxyfication
- update of beta / release notes
2014-12-12 11:10:17 -08:00

240 lines
11 KiB
C++

//
// Copyright 2014 DreamWorks Animation LLC.
//
// Licensed under the Apache License, Version 2.0 (the "Apache License")
// with the following modification; you may not use this file except in
// compliance with the Apache License and the following modification to it:
// Section 6. Trademarks. is deleted and replaced with:
//
// 6. Trademarks. This License does not grant permission to use the trade
// names, trademarks, service marks, or product names of the Licensor
// and its affiliates, except as required to comply with Section 4(c) of
// the License and to reproduce the content of the NOTICE file.
//
// You may obtain a copy of the Apache License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the Apache License with the above modification is
// distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the Apache License for the specific
// language governing permissions and limitations under the Apache License.
//
#ifndef SDC_CREASE_H
#define SDC_CREASE_H
#include "../version.h"
#include "../sdc/options.h"
namespace OpenSubdiv {
namespace OPENSUBDIV_VERSION {
namespace Sdc {
//
// Types, constants and utilities related to semi-sharp creasing -- whose implementation is
// independent of the subdivision scheme.
//
// Crease is intended to be a light-weight, trivially constructed class that computes
// crease-related properties. An instance of an Crease is defined with a set of options
// that include current and future variations that will impact computations involving
// sharpness values.
//
// We do not to use Neighborhoods here as input. Since their sharpness values are potentially
// not specified (and gathered on demand), and the methods here rely more on the sharpness
// values and less on the topology, we choose to work directly with the sharpness values for
// more flexibility. We also follow the trend of using primitive arrays in the interface.
//
// Note on the need for and use of sharpness values:
// In general, mask queries rely on the sharpness values. The common case of a smooth
// vertex, when known, avoids the need to inspect them, but unless the rules are well understood,
// users will be expected to provided them -- particularly when they expect the mask queries
// to do all of the work (just determining if a vertex is smooth will require inspection of
// incident edge sharpness).
// Mask queries will occassionally require the subdivided sharpness values around the
// child vertex. So users will be expected to either provide them up front when known, or to be
// gathered on demand. Any implementation of subdivision with creasing cannot avoid subdividing
// the sharpness values first, so keeping them available for re-use is a worthwhile consideration.
//
class Crease {
public:
//
// Constants and related queries of sharpness values:
//
static float const SHARPNESS_SMOOTH; // = 0.0f, do we really need this?
static float const SHARPNESS_INFINITE; // = 10.0f;
static bool IsSmooth(float sharpness) { return sharpness <= SHARPNESS_SMOOTH; }
static bool IsSharp(float sharpness) { return sharpness > SHARPNESS_SMOOTH; }
static bool IsInfinite(float sharpness) { return sharpness >= SHARPNESS_INFINITE; }
static bool IsSemiSharp(float sharpness) { return (SHARPNESS_SMOOTH < sharpness) && (sharpness < SHARPNESS_INFINITE); }
//
// Enum for the types of subdivision rules applied based on sharpness values (note these
// correspond to Hbr's vertex "mask"). The values are assigned to bit positions as it is
// useful to OR the corners of faces to quickly inspect its applicable rules.
//
enum Rule {
RULE_UNKNOWN = 0,
RULE_SMOOTH = (1 << 0),
RULE_DART = (1 << 1),
RULE_CREASE = (1 << 2),
RULE_CORNER = (1 << 3)
};
public:
Crease() : _options() { }
Crease(Options const& options) : _options(options) { }
~Crease() { }
//
// Considering labeling the current/default/normal creasing method as "simple" in contrast
// to all others that are "complex". The idea is that code can make certain assumptions
// and take some "simple" action in some cases to avoid the higher costs of dealing with
// more complex implementations.
//
bool IsUniform() const { return _options.GetCreasingMethod() == Options::CREASE_UNIFORM; }
//
// Optional sharp features:
// Since options treat certain topological features as infinitely sharp -- boundaries
// or nonmanifold features -- sharpness values should be adjust before use. The following
// methods will adjust specific according to the options applied.
//
float SharpenBoundaryEdge(float edgeSharpness) const;
float SharpenBoundaryVertex(float edgeSharpness) const;
float SharpenNonManifoldEdge(float edgeSharpness) const;
float SharpenNonManifoldVertex(float edgeSharpness) const;
//
// Sharpness subdivision:
// The simple case for computing a subdivided sharpness value is as follows:
// - Smooth edges or verts stay Smooth
// - Sharp edges or verts stay Sharp
// - semi-sharp edges or verts are decremented by 1.0
// but for Chaikin (and potentially future creasing schemes that improve upon it) the
// computation is more involved. In the case of edges in particular, the sharpness of a
// child edge is determined by the sharpness in the neighborhood of the end vertex
// corresponding to the child. For this reason, an alternative to subdividing sharpness
// that computes all child edges around a vertex is given.
//
float SubdivideUniformSharpness(float vertexOrEdgeSharpness) const;
float SubdivideVertexSharpness(float vertexSharpness) const;
float SubdivideEdgeSharpnessAtVertex(float edgeSharpness,
int incidentEdgeCountAtEndVertex,
float const* edgeSharpnessAroundEndVertex) const;
void SubdivideEdgeSharpnessesAroundVertex(int incidentEdgeCountAtVertex,
float const* incidentEdgeSharpnessAroundVertex,
float* childEdgesSharpnessAroundVertex) const;
//
// Rule determination:
// Mask queries do not require the Rule to be known, it can be determined from
// the information provided, but it is generally more efficient when the Rule is known
// and provided. In particular, the Smooth case dominates and is known to be applicable
// based on the origin of the vertex without inspection of sharpness.
//
Rule DetermineVertexVertexRule(float vertexSharpness,
int incidentEdgeCount,
float const* incidentEdgeSharpness) const;
Rule DetermineVertexVertexRule(float vertexSharpness,
int sharpEdgeCount) const;
//
// Transitional weighting:
// When the rules applicable to a parent vertex and its child differ, one or more
// sharpness values has "decayed" to zero. Both rules are then applicable and blended
// by a weight between 0 and 1 that reflects the transition. Most often this will be
// a single sharpness value that decays from within the interval [0,1] to zero -- and
// the weight to apply is exactly that sharpness value -- but more than one may decay,
// and values > 1 may also decay to 0 in a single step while others within [0,1] may
// remain > 0.
// So to properly determine a transitional weight, sharpness values for both the
// parent and child must be inspected, combined and clamped accordingly.
//
// Open questions:
// - does this method need to be public, or can it reside within the mask
// query classes? (though it would be the same for anything non-linear, so
// may be worth making a protected method somewhere)
// - does this need further consideration at an edge-vertex?
// - no, the edge-vertex case is far more trivial: one non-zero sharpness
// for the edge that decays to zero for one or both child edges -- the
// transitional weight is simply the edge sharpness (clamped to 1)
// ? why pass only the parent vertex sharpness...
// - because it is so trivial to compute the child vertex sharpness?
// - may be better off passing both parent and child for both vertex and edge
// just to be clear here.
//
float ComputeFractionalWeightAtVertex(float vertexSharpness,
float childVertexSharpness,
int incidentEdgeCount,
float const* incidentEdgeSharpness,
float const* childEdgesSharpness) const;
// Would these really help? Maybe only need Rules for the vertex-vertex case...
//
// Rule DetermineEdgeVertexRule(float parentEdgeSharpness) const;
// Rule DetermineEdgeVertexRule(float childEdge1Sharpness, float childEdge2Sharpness) const;
protected:
float decrementSharpness(float sharpness) const;
private:
Options _options;
};
// XXXX manuelk non-manifold is not implemented yet - return infintely sharp as default for now
//
// Non-trivial inline declarations:
//
inline float
Crease::SharpenBoundaryEdge(float edgeSharpness) const {
return (_options.GetVtxBoundaryInterpolation() != Options::VTX_BOUNDARY_NONE) ?
SHARPNESS_INFINITE : edgeSharpness;
}
inline float
Crease::SharpenBoundaryVertex(float vertexSharpness) const {
return (_options.GetVtxBoundaryInterpolation() == Options::VTX_BOUNDARY_EDGE_AND_CORNER) ?
SHARPNESS_INFINITE : vertexSharpness;
}
inline float
Crease::decrementSharpness(float sharpness) const {
if (IsSmooth(sharpness)) return Crease::SHARPNESS_SMOOTH; // redundant but most common
if (IsInfinite(sharpness)) return Crease::SHARPNESS_INFINITE;
if (sharpness > 1.0f) return (sharpness - 1.0f);
return Crease::SHARPNESS_SMOOTH;
}
inline float
Crease::SubdivideUniformSharpness(float vertexOrEdgeSharpness) const {
return decrementSharpness(vertexOrEdgeSharpness);
}
inline float
Crease::SubdivideVertexSharpness(float vertexSharpness) const {
return decrementSharpness(vertexSharpness);
}
} // end namespace sdc
} // end namespace OPENSUBDIV_VERSION
using namespace OPENSUBDIV_VERSION;
} // end namespace OpenSubdiv
#endif /* SDC_CREASE_H */