vpx/vp9/common/vp9_onyxc_int.h

363 lines
12 KiB
C
Raw Normal View History

2010-05-18 17:58:33 +02:00
/*
* Copyright (c) 2010 The WebM project authors. All Rights Reserved.
2010-05-18 17:58:33 +02:00
*
* Use of this source code is governed by a BSD-style license
* that can be found in the LICENSE file in the root of the source
* tree. An additional intellectual property rights grant can be found
* in the file PATENTS. All contributing project authors may
* be found in the AUTHORS file in the root of the source tree.
2010-05-18 17:58:33 +02:00
*/
#ifndef VP9_COMMON_VP9_ONYXC_INT_H_
#define VP9_COMMON_VP9_ONYXC_INT_H_
2010-05-18 17:58:33 +02:00
#include "vpx_config.h"
#include "vpx/internal/vpx_codec_internal.h"
#include "vp9_rtcd.h"
#include "vp9/common/vp9_loopfilter.h"
#include "vp9/common/vp9_entropymv.h"
#include "vp9/common/vp9_entropy.h"
#include "vp9/common/vp9_entropymode.h"
#include "vp9/common/vp9_quant_common.h"
#if CONFIG_POSTPROC
#include "vp9/common/vp9_postproc.h"
#endif
2010-05-18 17:58:33 +02:00
/* Create/destroy static data structures. */
// Define the number of candidate reference buffers.
#define NUM_REF_FRAMES 8
#define NUM_REF_FRAMES_LOG2 3
#define ALLOWED_REFS_PER_FRAME 3
// 1 scratch frame for the new frame, 3 for scaled references on the encoder
// TODO(jkoleszar): These 3 extra references could probably come from the
// normal reference pool.
#define NUM_YV12_BUFFERS (NUM_REF_FRAMES + 4)
2010-05-18 17:58:33 +02:00
#define NUM_FRAME_CONTEXTS_LOG2 2
#define NUM_FRAME_CONTEXTS (1 << NUM_FRAME_CONTEXTS_LOG2)
#define MAX_LAG_BUFFERS 25
typedef struct frame_contexts {
// y_mode, uv_mode, partition
vp9_prob y_mode_prob[BLOCK_SIZE_GROUPS][VP9_INTRA_MODES - 1];
vp9_prob uv_mode_prob[VP9_INTRA_MODES][VP9_INTRA_MODES - 1];
vp9_prob partition_prob[NUM_FRAME_TYPES][NUM_PARTITION_CONTEXTS]
[PARTITION_TYPES - 1];
vp9_prob pre_y_mode_prob[BLOCK_SIZE_GROUPS][VP9_INTRA_MODES - 1];
vp9_prob pre_uv_mode_prob[VP9_INTRA_MODES][VP9_INTRA_MODES - 1];
vp9_prob pre_partition_prob[NUM_PARTITION_CONTEXTS][PARTITION_TYPES - 1];
unsigned int y_mode_counts[BLOCK_SIZE_GROUPS][VP9_INTRA_MODES];
unsigned int uv_mode_counts[VP9_INTRA_MODES][VP9_INTRA_MODES];
unsigned int partition_counts[NUM_PARTITION_CONTEXTS][PARTITION_TYPES];
// coeff
vp9_coeff_probs_model coef_probs[TX_SIZE_MAX_SB][BLOCK_TYPES];
vp9_coeff_probs_model pre_coef_probs[TX_SIZE_MAX_SB][BLOCK_TYPES];
vp9_coeff_count_model coef_counts[TX_SIZE_MAX_SB][BLOCK_TYPES];
unsigned int eob_branch_counts[TX_SIZE_MAX_SB][BLOCK_TYPES][REF_TYPES]
[COEF_BANDS][PREV_COEF_CONTEXTS];
// switchable_interp
vp9_prob switchable_interp_prob[VP9_SWITCHABLE_FILTERS + 1]
[VP9_SWITCHABLE_FILTERS - 1];
vp9_prob pre_switchable_interp_prob[VP9_SWITCHABLE_FILTERS + 1]
[VP9_SWITCHABLE_FILTERS - 1];
unsigned int switchable_interp_count[VP9_SWITCHABLE_FILTERS + 1]
[VP9_SWITCHABLE_FILTERS];
// inter_mode
vp9_prob inter_mode_probs[INTER_MODE_CONTEXTS][VP9_INTER_MODES - 1];
vp9_prob pre_inter_mode_probs[INTER_MODE_CONTEXTS][VP9_INTER_MODES - 1];
unsigned int inter_mode_counts[INTER_MODE_CONTEXTS][VP9_INTER_MODES - 1][2];
// intra_inter, comp_inter, single_ref, comp_ref
vp9_prob intra_inter_prob[INTRA_INTER_CONTEXTS];
vp9_prob comp_inter_prob[COMP_INTER_CONTEXTS];
vp9_prob single_ref_prob[REF_CONTEXTS][2];
vp9_prob comp_ref_prob[REF_CONTEXTS];
vp9_prob pre_intra_inter_prob[INTRA_INTER_CONTEXTS];
vp9_prob pre_comp_inter_prob[COMP_INTER_CONTEXTS];
vp9_prob pre_single_ref_prob[REF_CONTEXTS][2];
vp9_prob pre_comp_ref_prob[REF_CONTEXTS];
unsigned int intra_inter_count[INTRA_INTER_CONTEXTS][2];
unsigned int comp_inter_count[COMP_INTER_CONTEXTS][2];
unsigned int single_ref_count[REF_CONTEXTS][2][2];
unsigned int comp_ref_count[REF_CONTEXTS][2];
// tx_probs
struct tx_probs tx_probs;
struct tx_probs pre_tx_probs;
struct tx_counts tx_counts;
// mbskip
vp9_prob mbskip_probs[MBSKIP_CONTEXTS];
vp9_prob pre_mbskip_probs[MBSKIP_CONTEXTS];
unsigned int mbskip_count[MBSKIP_CONTEXTS][2];
// mv
nmv_context nmvc;
nmv_context pre_nmvc;
nmv_context_counts NMVcount;
2010-05-18 17:58:33 +02:00
} FRAME_CONTEXT;
typedef enum {
SINGLE_PREDICTION_ONLY = 0,
COMP_PREDICTION_ONLY = 1,
HYBRID_PREDICTION = 2,
NB_PREDICTION_TYPES = 3,
} COMPPREDMODE_TYPE;
typedef enum {
ONLY_4X4 = 0,
ALLOW_8X8 = 1,
ALLOW_16X16 = 2,
32x32 transform for superblocks. This adds Debargha's DCT/DWT hybrid and a regular 32x32 DCT, and adds code all over the place to wrap that in the bitstream/encoder/decoder/RD. Some implementation notes (these probably need careful review): - token range is extended by 1 bit, since the value range out of this transform is [-16384,16383]. - the coefficients coming out of the FDCT are manually scaled back by 1 bit, or else they won't fit in int16_t (they are 17 bits). Because of this, the RD error scoring does not right-shift the MSE score by two (unlike for 4x4/8x8/16x16). - to compensate for this loss in precision, the quantizer is halved also. This is currently a little hacky. - FDCT and IDCT is double-only right now. Needs a fixed-point impl. - There are no default probabilities for the 32x32 transform yet; I'm simply using the 16x16 luma ones. A future commit will add newly generated probabilities for all transforms. - No ADST version. I don't think we'll add one for this level; if an ADST is desired, transform-size selection can scale back to 16x16 or lower, and use an ADST at that level. Additional notes specific to Debargha's DWT/DCT hybrid: - coefficient scale is different for the top/left 16x16 (DCT-over-DWT) block than for the rest (DWT pixel differences) of the block. Therefore, RD error scoring isn't easily scalable between coefficient and pixel domain. Thus, unfortunately, we need to compute the RD distortion in the pixel domain until we figure out how to scale these appropriately. Change-Id: I00386f20f35d7fabb19aba94c8162f8aee64ef2b
2012-12-07 23:45:05 +01:00
ALLOW_32X32 = 3,
TX_MODE_SELECT = 4,
NB_TXFM_MODES = 5,
Improved coding using 8x8 transform In summary, this commit encompasses a series of changes in attempt to improve the 8x8 transform based coding to help overall compression quality, please refer to the detailed commit history below for what are the rationale underly the series of changes: a. A frame level flag to indicate if 8x8 transform is used at all. b. 8x8 transform is not used for key frames and small image size. c. On inter coded frame, macroblocks using modes B_PRED, SPLIT_MV and I8X8_PRED are forced to using 4x4 transform based coding, the rest uses 8x8 transform based coding. d. Encoder and decoder has the same assumption on the relationship between prediction modes and transform size, therefore no signaling is encoded in bitstream. e. Mode decision process now calculate the rate and distortion scores using their respective transforms. Overall test results: 1. HD set http://www.corp.google.com/~yaowu/no_crawl/t8x8/HD_t8x8_20120206.html (avg psnr: 3.09% glb psnr: 3.22%, ssim: 3.90%) 2. Cif set: http://www.corp.google.com/~yaowu/no_crawl/t8x8/cif_t8x8_20120206.html (avg psnr: -0.03%, glb psnr: -0.02%, ssim: -0.04%) It should be noted here, as 8x8 transform coding itself is disabled for cif size clips, the 0.03% loss is purely from the 1 bit/frame flag overhead on if 8x8 transform is used or not for the frame. ---patch history for future reference--- Patch 1: this commit tries to select transform size based on macroblock prediction mode. If the size of a prediction mode is 16x16, then the macroblock is forced to use 8x8 transform. If the prediction mode is B_PRED, SPLITMV or I8X8_PRED, then the macroblock is forced to use 4x4 transform. Tests on the following HD clips showed mixed results: (all hd clips only used first 100 frames in the test) http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdmodebased8x8.html http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdmodebased8x8_log.html while the results are mixed and overall negative, it is interesting to see 8x8 helped a few of the clips. Patch 2: this patch tries to hard-wire selection of transform size based on prediction modes without using segmentation to signal the transform size. encoder and decoder both takes the same assumption that all macroblocks use 8x8 transform except when prediciton mode is B_PRED, I8X8_PRED or SPLITMV. Test results are as follows: http://www.corp.google.com/~yaowu/no_crawl/t8x8/cifmodebase8x8_0125.html http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdmodebased8x8_0125log.html Interestingly, by removing the overhead or coding the segmentation, the results on this limited HD set have turn positive on average. Patch 3: this patch disabled the usage of 8x8 transform on key frames, and kept the logic from patch 2 for inter frames only. test results on HD set turned decidedly positive with 8x8 transform enabled on inter frame with 16x16 prediction modes: (avg psnr: .81% glb psnr: .82 ssim: .55%) http://www.corp.google.com/~yaowu/no_crawl/t8x8/hdintermode8x8_0125.html results on cif set still negative overall Patch 4: continued from last patch, but now in mode decision process, the rate and distortion estimates are computed based on 8x8 transform results for MBs with modes associated with 8x8 transform. This patch also fixed a problem related to segment based eob coding when 8x8 transform is used. The patch significantly improved the results on HD clips: http://www.corp.google.com/~yaowu/no_crawl/t8x8/hd8x8RDintermode.html (avg psnr: 2.70% glb psnr: 2.76% ssim: 3.34%) results on cif also improved, though they are still negative compared to baseline that uses 4x4 transform only: http://www.corp.google.com/~yaowu/no_crawl/t8x8/cif8x8RDintermode.html (avg psnr: -.78% glb psnr: -.86% ssim: -.19%) Patch 5: This patch does 3 things: a. a bunch of decoder bug fixes, encodings and decodings were verified to have matched recon buffer on a number of encodes on cif size mobile and hd version of _pedestrian. b. the patch further improved the rate distortion calculation of MBS that use 8x8 transform. This provided some further gain on compression. c. the patch also got the experimental work SEG_LVL_EOB to work with 8x8 transformed macroblock, test results indicates it improves the cif set but hurt the HD set slightly. Tests results on HD clips: http://www.corp.google.com/~yaowu/no_crawl/t8x8/HD_t8x8_20120201.html (avg psnr: 3.19% glb psnr: 3.30% ssim: 3.93%) Test results on cif clips: http://www.corp.google.com/~yaowu/no_crawl/t8x8/cif_t8x8_20120201.html (avg psnr: -.47% glb psnr: -.51% ssim: +.28%) Patch 6: Added a frame level flag to indicate if 8x8 transform is allowed at all. temporarily the decision is based on frame size, can be optimized later one. This get the cif results to basically unchanged, with one bit per frame overhead on both cif and hd clips. Patch 8: Rebase and Merge to head by PGW. Fixed some suspect 4s that look like hey should be 64s in regard to segmented EOB. Perhaps #defines would be bette. Bulit and tested without T8x8 enabled and produces unchanged output. Patch 9: Corrected misalligned code/decode of "txfm_mode" bit. Limited testing for correct encode and decode with T8x8 configured on derf clips. Change-Id: I156e1405d25f81579d579dff8ab9af53944ec49c
2012-02-10 01:12:23 +01:00
} TXFM_MODE;
typedef struct VP9Common {
struct vpx_internal_error_info error;
2010-05-18 17:58:33 +02:00
DECLARE_ALIGNED(16, int16_t, y_dequant[QINDEX_RANGE][8]);
DECLARE_ALIGNED(16, int16_t, uv_dequant[QINDEX_RANGE][8]);
#if CONFIG_ALPHA
DECLARE_ALIGNED(16, int16_t, a_dequant[QINDEX_RANGE][8]);
#endif
2010-05-18 17:58:33 +02:00
int width;
int height;
int display_width;
int display_height;
int last_width;
int last_height;
2010-05-18 17:58:33 +02:00
// TODO(jkoleszar): this implies chroma ss right now, but could vary per
// plane. Revisit as part of the future change to YV12_BUFFER_CONFIG to
// support additional planes.
int subsampling_x;
int subsampling_y;
YUV_TYPE clr_type;
2010-05-18 17:58:33 +02:00
YV12_BUFFER_CONFIG *frame_to_show;
YV12_BUFFER_CONFIG yv12_fb[NUM_YV12_BUFFERS];
int fb_idx_ref_cnt[NUM_YV12_BUFFERS]; /* reference counts */
int ref_frame_map[NUM_REF_FRAMES]; /* maps fb_idx to reference slot */
// TODO(jkoleszar): could expand active_ref_idx to 4, with 0 as intra, and
// roll new_fb_idx into it.
// Each frame can reference ALLOWED_REFS_PER_FRAME buffers
int active_ref_idx[ALLOWED_REFS_PER_FRAME];
struct scale_factors active_ref_scale[ALLOWED_REFS_PER_FRAME];
int new_fb_idx;
YV12_BUFFER_CONFIG post_proc_buffer;
YV12_BUFFER_CONFIG temp_scale_frame;
2010-05-18 17:58:33 +02:00
FRAME_TYPE last_frame_type; /* Save last frame's frame type for motion search. */
FRAME_TYPE frame_type;
2010-05-18 17:58:33 +02:00
int show_frame;
int last_show_frame;
2010-05-18 17:58:33 +02:00
// Flag signaling that the frame is encoded using only INTRA modes.
int intra_only;
// Flag signaling that the frame context should be reset to default values.
// 0 or 1 implies don't reset, 2 reset just the context specified in the
// frame header, 3 reset all contexts.
int reset_frame_context;
int frame_flags;
// MBs, mb_rows/cols is in 16-pixel units; mi_rows/cols is in
// MODE_INFO (8-pixel) units.
int MBs;
int mb_rows, mi_rows;
int mb_cols, mi_cols;
int mode_info_stride;
2010-05-18 17:58:33 +02:00
/* profile settings */
TXFM_MODE txfm_mode;
2010-05-18 17:58:33 +02:00
int base_qindex;
int last_kf_gf_q; /* Q used on the last GF or KF */
2010-05-18 17:58:33 +02:00
int y_dc_delta_q;
int uv_dc_delta_q;
int uv_ac_delta_q;
#if CONFIG_ALPHA
int a_dc_delta_q;
int a_ac_delta_q;
#endif
2010-05-18 17:58:33 +02:00
unsigned int frames_since_golden;
unsigned int frames_till_alt_ref_frame;
2010-05-18 17:58:33 +02:00
/* We allocate a MODE_INFO struct for each macroblock, together with
an extra row on top and column on the left to simplify prediction. */
2010-05-18 17:58:33 +02:00
MODE_INFO *mip; /* Base of allocated array */
MODE_INFO *mi; /* Corresponds to upper left visible macroblock */
MODE_INFO *prev_mip; /* MODE_INFO array 'mip' from last decoded frame */
MODE_INFO *prev_mi; /* 'mi' from last frame (points into prev_mip) */
2010-05-18 17:58:33 +02:00
// Persistent mb segment id map used in prediction.
unsigned char *last_frame_seg_map;
INTERPOLATIONFILTERTYPE mcomp_filter_type;
loop_filter_info_n lf_info;
int filter_level;
int last_sharpness_level;
int sharpness_level;
2010-05-18 17:58:33 +02:00
int refresh_frame_context; /* Two state 0 = NO, 1 = YES */
2010-05-18 17:58:33 +02:00
int ref_frame_sign_bias[MAX_REF_FRAMES]; /* Two state 0, 1 */
2010-05-18 17:58:33 +02:00
/* Y,U,V */
ENTROPY_CONTEXT *above_context[MAX_MB_PLANE];
ENTROPY_CONTEXT left_context[MAX_MB_PLANE][16];
2010-05-18 17:58:33 +02:00
// partition contexts
PARTITION_CONTEXT *above_seg_context;
PARTITION_CONTEXT left_seg_context[8];
/* keyframe block modes are predicted by their above, left neighbors */
2010-05-18 17:58:33 +02:00
vp9_prob kf_y_mode_prob[VP9_INTRA_MODES]
[VP9_INTRA_MODES]
[VP9_INTRA_MODES - 1];
vp9_prob kf_uv_mode_prob[VP9_INTRA_MODES] [VP9_INTRA_MODES - 1];
// Context probabilities for reference frame prediction
int allow_comp_inter_inter;
MV_REFERENCE_FRAME comp_fixed_ref;
MV_REFERENCE_FRAME comp_var_ref[2];
COMPPREDMODE_TYPE comp_pred_mode;
FRAME_CONTEXT fc; /* this frame entropy */
FRAME_CONTEXT frame_contexts[NUM_FRAME_CONTEXTS];
unsigned int frame_context_idx; /* Context to use/update */
2010-05-18 17:58:33 +02:00
unsigned int current_video_frame;
int near_boffset[3];
int version;
2010-05-18 17:58:33 +02:00
#if CONFIG_POSTPROC
struct postproc_state postproc_state;
#endif
int error_resilient_mode;
int frame_parallel_decoding_mode;
[WIP] Add column-based tiling. This patch adds column-based tiling. The idea is to make each tile independently decodable (after reading the common frame header) and also independendly encodable (minus within-frame cost adjustments in the RD loop) to speed-up hardware & software en/decoders if they used multi-threading. Column-based tiling has the added advantage (over other tiling methods) that it minimizes realtime use-case latency, since all threads can start encoding data as soon as the first SB-row worth of data is available to the encoder. There is some test code that does random tile ordering in the decoder, to confirm that each tile is indeed independently decodable from other tiles in the same frame. At tile edges, all contexts assume default values (i.e. 0, 0 motion vector, no coefficients, DC intra4x4 mode), and motion vector search and ordering do not cross tiles in the same frame. t log Tile independence is not maintained between frames ATM, i.e. tile 0 of frame 1 is free to use motion vectors that point into any tile of frame 0. We support 1 (i.e. no tiling), 2 or 4 column-tiles. The loopfilter crosses tile boundaries. I discussed this briefly with Aki and he says that's OK. An in-loop loopfilter would need to do some sync between tile threads, but that shouldn't be a big issue. Resuls: with tiling disabled, we go up slightly because of improved edge use in the intra4x4 prediction. With 2 tiles, we lose about ~1% on derf, ~0.35% on HD and ~0.55% on STD/HD. With 4 tiles, we lose another ~1.5% on derf ~0.77% on HD and ~0.85% on STD/HD. Most of this loss is concentrated in the low-bitrate end of clips, and most of it is because of the loss of edges at tile boundaries and the resulting loss of intra predictors. TODO: - more tiles (perhaps allow row-based tiling also, and max. 8 tiles)? - maybe optionally (for EC purposes), motion vectors themselves should not cross tile edges, or we should emulate such borders as if they were off-frame, to limit error propagation to within one tile only. This doesn't have to be the default behaviour but could be an optional bitstream flag. Change-Id: I5951c3a0742a767b20bc9fb5af685d9892c2c96f
2013-02-01 18:35:28 +01:00
int tile_columns, log2_tile_columns;
int cur_tile_mi_col_start, cur_tile_mi_col_end, cur_tile_col_idx;
int tile_rows, log2_tile_rows;
int cur_tile_mi_row_start, cur_tile_mi_row_end, cur_tile_row_idx;
} VP9_COMMON;
2010-05-18 17:58:33 +02:00
static int get_free_fb(VP9_COMMON *cm) {
int i;
for (i = 0; i < NUM_YV12_BUFFERS; i++)
if (cm->fb_idx_ref_cnt[i] == 0)
break;
assert(i < NUM_YV12_BUFFERS);
cm->fb_idx_ref_cnt[i] = 1;
return i;
}
static void ref_cnt_fb(int *buf, int *idx, int new_idx) {
if (buf[*idx] > 0)
buf[*idx]--;
*idx = new_idx;
buf[new_idx]++;
}
static int mi_cols_aligned_to_sb(int n_mis) {
return ALIGN_POWER_OF_TWO(n_mis, LOG2_MI_BLOCK_SIZE);
}
static INLINE void set_partition_seg_context(VP9_COMMON *cm, MACROBLOCKD *xd,
int mi_row, int mi_col) {
xd->above_seg_context = cm->above_seg_context + mi_col;
xd->left_seg_context = cm->left_seg_context + (mi_row & MI_MASK);
}
static int check_bsize_coverage(VP9_COMMON *cm, MACROBLOCKD *xd,
int mi_row, int mi_col,
BLOCK_SIZE_TYPE bsize) {
int bsl = mi_width_log2(bsize), bs = 1 << bsl;
int ms = bs / 2;
if ((mi_row + ms < cm->mi_rows) && (mi_col + ms < cm->mi_cols))
return 0;
// frame width/height are multiples of 8, hence 8x8 block should always
// pass the above check
assert(bsize > BLOCK_SIZE_SB8X8);
// return the node index in the prob tree for binary coding
// only allow horizontal/split partition types
if ((mi_col + ms < cm->mi_cols) && (mi_row + ms >= cm->mi_rows))
return 1;
// only allow vertical/split partition types
if ((mi_row + ms < cm->mi_rows) && (mi_col + ms >= cm->mi_cols))
return 2;
return -1;
}
static void set_mi_row_col(VP9_COMMON *cm, MACROBLOCKD *xd,
int mi_row, int bh,
int mi_col, int bw) {
xd->mb_to_top_edge = -((mi_row * MI_SIZE) << 3);
xd->mb_to_bottom_edge = ((cm->mi_rows - bh - mi_row) * MI_SIZE) << 3;
xd->mb_to_left_edge = -((mi_col * MI_SIZE) << 3);
xd->mb_to_right_edge = ((cm->mi_cols - bw - mi_col) * MI_SIZE) << 3;
// Are edges available for intra prediction?
xd->up_available = (mi_row != 0);
xd->left_available = (mi_col > cm->cur_tile_mi_col_start);
xd->right_available = (mi_col + bw < cm->cur_tile_mi_col_end);
}
static int get_mi_row(const MACROBLOCKD *xd) {
return ((-xd->mb_to_top_edge) >> (3 + LOG2_MI_SIZE));
}
static int get_mi_col(const MACROBLOCKD *xd) {
return ((-xd->mb_to_left_edge) >> (3 + LOG2_MI_SIZE));
}
static int get_token_alloc(int mb_rows, int mb_cols) {
return mb_rows * mb_cols * (48 * 16 + 4);
}
static void set_prev_mi(VP9_COMMON *cm) {
const int use_prev_in_find_mv_refs = cm->width == cm->last_width &&
cm->height == cm->last_height &&
!cm->error_resilient_mode &&
!cm->intra_only &&
cm->last_show_frame;
// Special case: set prev_mi to NULL when the previous mode info
// context cannot be used.
cm->prev_mi = use_prev_in_find_mv_refs ?
cm->prev_mip + cm->mode_info_stride + 1 : NULL;
}
#endif // VP9_COMMON_VP9_ONYXC_INT_H_