FFmpeg
rv34dsp.c
Go to the documentation of this file.
1 /*
2  * RV30/40 decoder common dsp functions
3  * Copyright (c) 2007 Mike Melanson, Konstantin Shishkov
4  * Copyright (c) 2011 Janne Grunau
5  *
6  * This file is part of FFmpeg.
7  *
8  * FFmpeg is free software; you can redistribute it and/or
9  * modify it under the terms of the GNU Lesser General Public
10  * License as published by the Free Software Foundation; either
11  * version 2.1 of the License, or (at your option) any later version.
12  *
13  * FFmpeg is distributed in the hope that it will be useful,
14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16  * Lesser General Public License for more details.
17  *
18  * You should have received a copy of the GNU Lesser General Public
19  * License along with FFmpeg; if not, write to the Free Software
20  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21  */
22 
23 /**
24  * @file
25  * RV30/40 decoder common dsp functions
26  */
27 
28 #include "rv34dsp.h"
29 #include "libavutil/common.h"
30 
31 /**
32  * @name RV30/40 inverse transform functions
33  * @{
34  */
35 
36 static av_always_inline void rv34_row_transform(int temp[16], int16_t *block)
37 {
38  int i;
39 
40  for(i = 0; i < 4; i++){
41  const int z0 = 13*(block[i+4*0] + block[i+4*2]);
42  const int z1 = 13*(block[i+4*0] - block[i+4*2]);
43  const int z2 = 7* block[i+4*1] - 17*block[i+4*3];
44  const int z3 = 17* block[i+4*1] + 7*block[i+4*3];
45 
46  temp[4*i+0] = z0 + z3;
47  temp[4*i+1] = z1 + z2;
48  temp[4*i+2] = z1 - z2;
49  temp[4*i+3] = z0 - z3;
50  }
51 }
52 
53 /**
54  * Real Video 3.0/4.0 inverse transform + sample reconstruction
55  * Code is almost the same as in SVQ3, only scaling is different.
56  */
57 static void rv34_idct_add_c(uint8_t *dst, ptrdiff_t stride, int16_t *block){
58  int temp[16];
59  int i;
60 
62  memset(block, 0, 16*sizeof(int16_t));
63 
64  for(i = 0; i < 4; i++){
65  const int z0 = 13*(temp[4*0+i] + temp[4*2+i]) + 0x200;
66  const int z1 = 13*(temp[4*0+i] - temp[4*2+i]) + 0x200;
67  const int z2 = 7* temp[4*1+i] - 17*temp[4*3+i];
68  const int z3 = 17* temp[4*1+i] + 7*temp[4*3+i];
69 
70  dst[0] = av_clip_uint8( dst[0] + ( (z0 + z3) >> 10 ) );
71  dst[1] = av_clip_uint8( dst[1] + ( (z1 + z2) >> 10 ) );
72  dst[2] = av_clip_uint8( dst[2] + ( (z1 - z2) >> 10 ) );
73  dst[3] = av_clip_uint8( dst[3] + ( (z0 - z3) >> 10 ) );
74 
75  dst += stride;
76  }
77 }
78 
79 /**
80  * RealVideo 3.0/4.0 inverse transform for DC block
81  *
82  * Code is almost the same as rv34_inv_transform()
83  * but final coefficients are multiplied by 1.5 and have no rounding.
84  */
85 static void rv34_inv_transform_noround_c(int16_t *block){
86  int temp[16];
87  int i;
88 
90 
91  for(i = 0; i < 4; i++){
92  const int z0 = 39*(temp[4*0+i] + temp[4*2+i]);
93  const int z1 = 39*(temp[4*0+i] - temp[4*2+i]);
94  const int z2 = 21* temp[4*1+i] - 51*temp[4*3+i];
95  const int z3 = 51* temp[4*1+i] + 21*temp[4*3+i];
96 
97  block[i*4+0] = (z0 + z3) >> 11;
98  block[i*4+1] = (z1 + z2) >> 11;
99  block[i*4+2] = (z1 - z2) >> 11;
100  block[i*4+3] = (z0 - z3) >> 11;
101  }
102 }
103 
104 static void rv34_idct_dc_add_c(uint8_t *dst, ptrdiff_t stride, int dc)
105 {
106  int i, j;
107 
108  dc = (13*13*dc + 0x200) >> 10;
109  for (i = 0; i < 4; i++)
110  {
111  for (j = 0; j < 4; j++)
112  dst[j] = av_clip_uint8( dst[j] + dc );
113 
114  dst += stride;
115  }
116 }
117 
119 {
120  int16_t dc = (13 * 13 * 3 * block[0]) >> 11;
121  int i, j;
122 
123  for (i = 0; i < 4; i++, block += 4)
124  for (j = 0; j < 4; j++)
125  block[j] = dc;
126 }
127 
128 /** @} */ // transform
129 
130 
132 {
133  c->rv34_inv_transform = rv34_inv_transform_noround_c;
134  c->rv34_inv_transform_dc = rv34_inv_transform_dc_noround_c;
135 
136  c->rv34_idct_add = rv34_idct_add_c;
137  c->rv34_idct_dc_add = rv34_idct_dc_add_c;
138 
139  if (ARCH_ARM)
141  if (ARCH_X86)
143 }
stride
int stride
Definition: mace.c:144
rv34_idct_dc_add_c
static void rv34_idct_dc_add_c(uint8_t *dst, ptrdiff_t stride, int dc)
Definition: rv34dsp.c:104
rv34_inv_transform_noround_c
static void rv34_inv_transform_noround_c(int16_t *block)
RealVideo 3.0/4.0 inverse transform for DC block.
Definition: rv34dsp.c:85
av_cold
#define av_cold
Definition: attributes.h:90
rv34_row_transform
static av_always_inline void rv34_row_transform(int temp[16], int16_t *block)
Definition: rv34dsp.c:36
ff_rv34dsp_init
av_cold void ff_rv34dsp_init(RV34DSPContext *c)
Definition: rv34dsp.c:131
rv34_inv_transform_dc_noround_c
static void rv34_inv_transform_dc_noround_c(int16_t *block)
Definition: rv34dsp.c:118
c
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
Definition: undefined.txt:32
RV34DSPContext
Definition: rv34dsp.h:57
dc
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2] ... the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so ...,+,-,+,-,+,+,-,+,-,+,... hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32 - hcoeff[1] - hcoeff[2] - ... a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2} an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||......... intra?||||:Block01 :yes no||||:Block02 :....... ..........||||:Block03 ::y DC ::ref index:||||:Block04 ::cb DC ::motion x :||||......... :cr DC ::motion y :||||....... ..........|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------ ------------ ------------|||Y subbands||Cb subbands||Cr subbands||||--- ---||--- ---||--- ---|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||--- ---||--- ---||--- ---||||--- ---||--- ---||--- ---|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------ ------------ ------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction ------------|\ Dequantization ------------------- \||Reference frames|\ IDWT|------- -------|Motion \|||Frame 0||Frame 1||Compensation . OBMC v -------|------- -------|--------------. \------> Frame n output Frame Frame<----------------------------------/|...|------------------- Range Coder:============Binary Range Coder:------------------- The implemented range coder is an adapted version based upon "Range encoding: an algorithm for removing redundancy from a digitised message." by G. N. N. Martin. The symbols encoded by the Snow range coder are bits(0|1). The associated probabilities are not fix but change depending on the symbol mix seen so far. bit seen|new state ---------+----------------------------------------------- 0|256 - state_transition_table[256 - old_state];1|state_transition_table[old_state];state_transition_table={ 0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:------------------------- FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1. the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled top and top right vectors is used as motion vector prediction the used motion vector is the sum of the predictor and(mvx_diff, mvy_diff) *mv_scale Intra DC Prediction block[y][x] dc[1]
Definition: snow.txt:400
rv34dsp.h
i
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:271
ff_rv34dsp_init_arm
av_cold void ff_rv34dsp_init_arm(RV34DSPContext *c)
Definition: rv34dsp_init_arm.c:35
common.h
av_always_inline
#define av_always_inline
Definition: attributes.h:49
ff_rv34dsp_init_x86
void ff_rv34dsp_init_x86(RV34DSPContext *c)
Definition: rv34dsp_init.c:34
rv34_idct_add_c
static void rv34_idct_add_c(uint8_t *dst, ptrdiff_t stride, int16_t *block)
Real Video 3.0/4.0 inverse transform + sample reconstruction Code is almost the same as in SVQ3,...
Definition: rv34dsp.c:57
temp
else temp
Definition: vf_mcdeint.c:248
av_clip_uint8
#define av_clip_uint8
Definition: common.h:102
block
The exact code depends on how similar the blocks are and how related they are to the block
Definition: filter_design.txt:207