Skip to content

O
opencv
Commit 07729e39 authored by Alexander Alekhin's avatar Alexander Alekhin
Browse files
Options

imgproc(resize): avoid unnecessary type conversions

parent 5e2bcc91
Show whitespace changes
Inline Side-by-side
Showing with 39 additions and 68 deletions
modules/imgproc/src/resize.cpp
View file @ 07729e39
...@@ -1109,23 +1109,29 @@ resizeNN( const Mat& src, Mat& dst, double fx, double fy ) ...@@ -1109,23 +1109,29 @@ resizeNN( const Mat& src, Mat& dst, double fx, double fy )
struct VResizeNoVec struct VResizeNoVec
{ {
int operator()(const uchar**, uchar*, const uchar*, int ) const { return 0; } template<typename WT, typename T, typename BT>
int operator()(const WT**, T*, const BT*, int ) const
{
return 0;
}
}; };
struct HResizeNoVec struct HResizeNoVec
{ {
int operator()(const uchar**, uchar**, int, const int*, template<typename T, typename WT, typename AT> inline
const uchar*, int, int, int, int, int) const { return 0; } int operator()(const T**, WT**, int, const int*,
const AT*, int, int, int, int, int) const
{
return 0;
}
}; };
#if CV_SIMD #if CV_SIMD
struct VResizeLinearVec_32s8u struct VResizeLinearVec_32s8u
{ {
int operator()(const uchar** _src, uchar* dst, const uchar* _beta, int width ) const int operator()(const int** src, uchar* dst, const short* beta, int width) const
{ {
const int** src = (const int**)_src;
const short* beta = (const short*)_beta;
const int *S0 = src[0], *S1 = src[1]; const int *S0 = src[0], *S1 = src[1];
int x = 0; int x = 0;
v_int16 b0 = vx_setall_s16(beta[0]), b1 = vx_setall_s16(beta[1]); v_int16 b0 = vx_setall_s16(beta[0]), b1 = vx_setall_s16(beta[1]);
...@@ -1153,12 +1159,9 @@ struct VResizeLinearVec_32s8u ...@@ -1153,12 +1159,9 @@ struct VResizeLinearVec_32s8u
struct VResizeLinearVec_32f16u struct VResizeLinearVec_32f16u
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, ushort* dst, const float* beta, int width) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1]; const float *S0 = src[0], *S1 = src[1];
ushort* dst = (ushort*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]); v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]);
...@@ -1183,12 +1186,9 @@ struct VResizeLinearVec_32f16u ...@@ -1183,12 +1186,9 @@ struct VResizeLinearVec_32f16u
struct VResizeLinearVec_32f16s struct VResizeLinearVec_32f16s
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, short* dst, const float* beta, int width) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1]; const float *S0 = src[0], *S1 = src[1];
short* dst = (short*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]); v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]);
...@@ -1213,12 +1213,9 @@ struct VResizeLinearVec_32f16s ...@@ -1213,12 +1213,9 @@ struct VResizeLinearVec_32f16s
struct VResizeLinearVec_32f struct VResizeLinearVec_32f
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, float* dst, const float* beta, int width) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1]; const float *S0 = src[0], *S1 = src[1];
float* dst = (float*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]); v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]);
...@@ -1237,10 +1234,8 @@ struct VResizeLinearVec_32f ...@@ -1237,10 +1234,8 @@ struct VResizeLinearVec_32f
struct VResizeCubicVec_32s8u struct VResizeCubicVec_32s8u
{ {
int operator()(const uchar** _src, uchar* dst, const uchar* _beta, int width ) const int operator()(const int** src, uchar* dst, const short* beta, int width) const
{ {
const int** src = (const int**)_src;
const short* beta = (const short*)_beta;
const int *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3]; const int *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3];
int x = 0; int x = 0;
float scale = 1.f/(INTER_RESIZE_COEF_SCALE*INTER_RESIZE_COEF_SCALE); float scale = 1.f/(INTER_RESIZE_COEF_SCALE*INTER_RESIZE_COEF_SCALE);
...@@ -1274,12 +1269,9 @@ struct VResizeCubicVec_32s8u ...@@ -1274,12 +1269,9 @@ struct VResizeCubicVec_32s8u
struct VResizeCubicVec_32f16u struct VResizeCubicVec_32f16u
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, ushort* dst, const float* beta, int width) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3]; const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3];
ushort* dst = (ushort*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]), v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]),
b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]); b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]);
...@@ -1300,12 +1292,9 @@ struct VResizeCubicVec_32f16u ...@@ -1300,12 +1292,9 @@ struct VResizeCubicVec_32f16u
struct VResizeCubicVec_32f16s struct VResizeCubicVec_32f16s
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, short* dst, const float* beta, int width) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3]; const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3];
short* dst = (short*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]), v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]),
b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]); b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]);
...@@ -1326,12 +1315,9 @@ struct VResizeCubicVec_32f16s ...@@ -1326,12 +1315,9 @@ struct VResizeCubicVec_32f16s
struct VResizeCubicVec_32f struct VResizeCubicVec_32f
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, float* dst, const float* beta, int width) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3]; const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3];
float* dst = (float*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]), v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]),
b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]); b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]);
...@@ -1351,10 +1337,12 @@ struct VResizeCubicVec_32f ...@@ -1351,10 +1337,12 @@ struct VResizeCubicVec_32f
struct VResizeLanczos4Vec_32f16u struct VResizeLanczos4Vec_32f16u
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, ushort* dst, const float* beta, int width) const
{ {
if (CV_CPU_HAS_SUPPORT_SSE4_1) return opt_SSE4_1::VResizeLanczos4Vec_32f16u_SSE41(_src, _dst, _beta, width); if (CV_CPU_HAS_SUPPORT_SSE4_1)
else return 0; return opt_SSE4_1::VResizeLanczos4Vec_32f16u_SSE41(src, dst, beta, width);
else
return 0;
} }
}; };
...@@ -1362,13 +1350,10 @@ struct VResizeLanczos4Vec_32f16u ...@@ -1362,13 +1350,10 @@ struct VResizeLanczos4Vec_32f16u
struct VResizeLanczos4Vec_32f16u struct VResizeLanczos4Vec_32f16u
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, ushort* dst, const float* beta, int width ) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3], const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3],
*S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7]; *S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7];
ushort * dst = (ushort*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]), v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]),
b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]), b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]),
...@@ -1401,13 +1386,10 @@ struct VResizeLanczos4Vec_32f16u ...@@ -1401,13 +1386,10 @@ struct VResizeLanczos4Vec_32f16u
struct VResizeLanczos4Vec_32f16s struct VResizeLanczos4Vec_32f16s
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, short* dst, const float* beta, int width ) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3], const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3],
*S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7]; *S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7];
short * dst = (short*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]), v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]),
b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]), b2 = vx_setall_f32(beta[2]), b3 = vx_setall_f32(beta[3]),
...@@ -1438,13 +1420,10 @@ struct VResizeLanczos4Vec_32f16s ...@@ -1438,13 +1420,10 @@ struct VResizeLanczos4Vec_32f16s
struct VResizeLanczos4Vec_32f struct VResizeLanczos4Vec_32f
{ {
int operator()(const uchar** _src, uchar* _dst, const uchar* _beta, int width ) const int operator()(const float** src, float* dst, const float* beta, int width ) const
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3], const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3],
*S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7]; *S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7];
float* dst = (float*)_dst;
int x = 0; int x = 0;
v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]), v_float32 b0 = vx_setall_f32(beta[0]), b1 = vx_setall_f32(beta[1]),
...@@ -1489,12 +1468,9 @@ typedef VResizeNoVec VResizeLanczos4Vec_32f; ...@@ -1489,12 +1468,9 @@ typedef VResizeNoVec VResizeLanczos4Vec_32f;
template<typename ST, typename DT, typename AT, typename DVT> template<typename ST, typename DT, typename AT, typename DVT>
struct HResizeLinearVec_X4 struct HResizeLinearVec_X4
{ {
int operator()(const uchar** _src, uchar** _dst, int count, const int* xofs, int operator()(const ST** src, DT** dst, int count, const int* xofs,
const uchar* _alpha, int, int, int cn, int, int xmax) const const AT* alpha, int, int, int cn, int, int xmax) const
{ {
const ST **src = (const ST**)_src;
const AT *alpha = (const AT*)_alpha;
DT **dst = (DT**)_dst;
const int nlanes = 4; const int nlanes = 4;
const int len0 = xmax & -nlanes; const int len0 = xmax & -nlanes;
int dx = 0, k = 0; int dx = 0, k = 0;
...@@ -1549,11 +1525,9 @@ struct HResizeLinearVec_X4 ...@@ -1549,11 +1525,9 @@ struct HResizeLinearVec_X4
struct HResizeLinearVecU8_X4 struct HResizeLinearVecU8_X4
{ {
int operator()(const uchar** src, uchar** _dst, int count, const int* xofs, int operator()(const uchar** src, int** dst, int count, const int* xofs,
const uchar* _alpha, int smax, int, int cn, int, int xmax) const const short* alpha/*[xmax]*/, int smax, int /*dmax*/, int cn, int /*xmin*/, int xmax) const
{ {
const short *alpha = (const short*)_alpha;
int **dst = (int**)_dst;
int dx = 0, k = 0; int dx = 0, k = 0;
if(cn == 1) if(cn == 1)
...@@ -1827,8 +1801,8 @@ struct HResizeLinear ...@@ -1827,8 +1801,8 @@ struct HResizeLinear
int dx, k; int dx, k;
VecOp vecOp; VecOp vecOp;
int dx0 = vecOp((const uchar**)src, (uchar**)dst, count, int dx0 = vecOp(src, dst, count,
xofs, (const uchar*)alpha, swidth, dwidth, cn, xmin, xmax ); xofs, alpha, swidth, dwidth, cn, xmin, xmax );
for( k = 0; k <= count - 2; k+=2 ) for( k = 0; k <= count - 2; k+=2 )
{ {
...@@ -1881,7 +1855,7 @@ struct VResizeLinear ...@@ -1881,7 +1855,7 @@ struct VResizeLinear
CastOp castOp; CastOp castOp;
VecOp vecOp; VecOp vecOp;
int x = vecOp((const uchar**)src, (uchar*)dst, (const uchar*)beta, width); int x = vecOp(src, dst, beta, width);
#if CV_ENABLE_UNROLLED #if CV_ENABLE_UNROLLED
for( ; x <= width - 4; x += 4 ) for( ; x <= width - 4; x += 4 )
{ {
...@@ -1912,7 +1886,7 @@ struct VResizeLinear<uchar, int, short, FixedPtCast<int, uchar, INTER_RESIZE_COE ...@@ -1912,7 +1886,7 @@ struct VResizeLinear<uchar, int, short, FixedPtCast<int, uchar, INTER_RESIZE_COE
const buf_type *S0 = src[0], *S1 = src[1]; const buf_type *S0 = src[0], *S1 = src[1];
VResizeLinearVec_32s8u vecOp; VResizeLinearVec_32s8u vecOp;
int x = vecOp((const uchar**)src, (uchar*)dst, (const uchar*)beta, width); int x = vecOp(src, dst, beta, width);
#if CV_ENABLE_UNROLLED #if CV_ENABLE_UNROLLED
for( ; x <= width - 4; x += 4 ) for( ; x <= width - 4; x += 4 )
{ {
...@@ -1994,7 +1968,7 @@ struct VResizeCubic ...@@ -1994,7 +1968,7 @@ struct VResizeCubic
CastOp castOp; CastOp castOp;
VecOp vecOp; VecOp vecOp;
int x = vecOp((const uchar**)src, (uchar*)dst, (const uchar*)beta, width); int x = vecOp(src, dst, beta, width);
for( ; x < width; x++ ) for( ; x < width; x++ )
dst[x] = castOp(S0[x]*b0 + S1[x]*b1 + S2[x]*b2 + S3[x]*b3); dst[x] = castOp(S0[x]*b0 + S1[x]*b1 + S2[x]*b2 + S3[x]*b3);
} }
...@@ -2066,7 +2040,7 @@ struct VResizeLanczos4 ...@@ -2066,7 +2040,7 @@ struct VResizeLanczos4
{ {
CastOp castOp; CastOp castOp;
VecOp vecOp; VecOp vecOp;
int x = vecOp((const uchar**)src, (uchar*)dst, (const uchar*)beta, width); int x = vecOp(src, dst, beta, width);
#if CV_ENABLE_UNROLLED #if CV_ENABLE_UNROLLED
for( ; x <= width - 4; x += 4 ) for( ; x <= width - 4; x += 4 )
{ {
......
modules/imgproc/src/resize.hpp
View file @ 07729e39
...@@ -67,7 +67,7 @@ namespace opt_SSE4_1 ...@@ -67,7 +67,7 @@ namespace opt_SSE4_1
void resizeNN2_SSE4_1(const Range&, const Mat&, Mat&, int*, int, double); void resizeNN2_SSE4_1(const Range&, const Mat&, Mat&, int*, int, double);
void resizeNN4_SSE4_1(const Range&, const Mat&, Mat&, int*, int, double); void resizeNN4_SSE4_1(const Range&, const Mat&, Mat&, int*, int, double);
int VResizeLanczos4Vec_32f16u_SSE41(const uchar** _src, uchar* _dst, const uchar* _beta, int width); int VResizeLanczos4Vec_32f16u_SSE41(const float** src, ushort* dst, const float* beta, int width);
#endif #endif
} }
} }
......
modules/imgproc/src/resize.sse4_1.cpp
View file @ 07729e39
...@@ -186,13 +186,10 @@ void resizeNN4_SSE4_1(const Range& range, const Mat& src, Mat &dst, int *x_ofs, ...@@ -186,13 +186,10 @@ void resizeNN4_SSE4_1(const Range& range, const Mat& src, Mat &dst, int *x_ofs,
parallel_for_(range, invoker, dst.total() / (double)(1 << 16)); parallel_for_(range, invoker, dst.total() / (double)(1 << 16));
} }
int VResizeLanczos4Vec_32f16u_SSE41(const uchar** _src, uchar* _dst, const uchar* _beta, int width) int VResizeLanczos4Vec_32f16u_SSE41(const float** src, ushort* dst, const float* beta, int width)
{ {
const float** src = (const float**)_src;
const float* beta = (const float*)_beta;
const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3], const float *S0 = src[0], *S1 = src[1], *S2 = src[2], *S3 = src[3],
*S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7]; *S4 = src[4], *S5 = src[5], *S6 = src[6], *S7 = src[7];
short * dst = (short*)_dst;
int x = 0; int x = 0;
__m128 v_b0 = _mm_set1_ps(beta[0]), v_b1 = _mm_set1_ps(beta[1]), __m128 v_b0 = _mm_set1_ps(beta[0]), v_b1 = _mm_set1_ps(beta[1]),
v_b2 = _mm_set1_ps(beta[2]), v_b3 = _mm_set1_ps(beta[3]), v_b2 = _mm_set1_ps(beta[2]), v_b3 = _mm_set1_ps(beta[3]),
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment