diff options
Diffstat (limited to 'media/libaom/src/av1/encoder/speed_features.h')
| -rw-r--r-- | media/libaom/src/av1/encoder/speed_features.h | 1004 |
1 files changed, 735 insertions, 269 deletions
diff --git a/media/libaom/src/av1/encoder/speed_features.h b/media/libaom/src/av1/encoder/speed_features.h index 41013b2e79..d12c3c02e1 100644 --- a/media/libaom/src/av1/encoder/speed_features.h +++ b/media/libaom/src/av1/encoder/speed_features.h @@ -18,6 +18,38 @@ extern "C" { #endif +#define MAX_MESH_STEP 4 + +typedef struct MESH_PATTERN { + int range; + int interval; +} MESH_PATTERN; + +enum { + GM_FULL_SEARCH, + GM_REDUCED_REF_SEARCH_SKIP_L2_L3, + GM_REDUCED_REF_SEARCH_SKIP_L2_L3_ARF2, + GM_DISABLE_SEARCH +} UENUM1BYTE(GM_SEARCH_TYPE); + +enum { + GM_ERRORADV_TR_0, + GM_ERRORADV_TR_1, + GM_ERRORADV_TR_2, + GM_ERRORADV_TR_TYPES, +} UENUM1BYTE(GM_ERRORADV_TYPE); + +enum { + FULL_TXFM_RD, + LOW_TXFM_RD, +} UENUM1BYTE(TXFM_RD_MODEL); + +enum { + DIST_WTD_COMP_ENABLED, + DIST_WTD_COMP_SKIP_MV_SEARCH, + DIST_WTD_COMP_DISABLED, +} UENUM1BYTE(DIST_WTD_COMP_FLAG); + enum { INTRA_ALL = (1 << DC_PRED) | (1 << V_PRED) | (1 << H_PRED) | (1 << D45_PRED) | (1 << D135_PRED) | (1 << D113_PRED) | (1 << D157_PRED) | @@ -73,7 +105,7 @@ enum { (1 << THR_ALTR) | (1 << THR_GOLD) }; -typedef enum { +enum { TXFM_CODING_SF = 1, INTER_PRED_SF = 2, INTRA_PRED_SF = 4, @@ -82,19 +114,9 @@ typedef enum { RD_SKIP_SF = 32, RESERVE_2_SF = 64, RESERVE_3_SF = 128, -} DEV_SPEED_FEATURES; - -typedef enum { - DIAMOND = 0, - NSTEP = 1, - HEX = 2, - BIGDIA = 3, - SQUARE = 4, - FAST_HEX = 5, - FAST_DIAMOND = 6 -} SEARCH_METHODS; - -typedef enum { +} UENUM1BYTE(DEV_SPEED_FEATURES); + +enum { // No recode. DISALLOW_RECODE = 0, // Allow recode for KF and exceeding maximum frame bandwidth. @@ -103,39 +125,44 @@ typedef enum { ALLOW_RECODE_KFARFGF = 2, // Allow recode for all frames based on bitrate constraints. ALLOW_RECODE = 3, -} RECODE_LOOP_TYPE; +} UENUM1BYTE(RECODE_LOOP_TYPE); -typedef enum { +enum { SUBPEL_TREE = 0, SUBPEL_TREE_PRUNED = 1, // Prunes 1/2-pel searches SUBPEL_TREE_PRUNED_MORE = 2, // Prunes 1/2-pel searches more aggressively SUBPEL_TREE_PRUNED_EVENMORE = 3, // Prunes 1/2- and 1/4-pel searches // Other methods to come -} SUBPEL_SEARCH_METHODS; +} UENUM1BYTE(SUBPEL_SEARCH_METHODS); -typedef enum { +enum { USE_FULL_RD = 0, USE_FAST_RD, USE_LARGESTALL, -} TX_SIZE_SEARCH_METHOD; - -typedef enum { - NOT_IN_USE = 0, - RELAXED_NEIGHBORING_MIN_MAX = 1 -} AUTO_MIN_MAX_MODE; +} UENUM1BYTE(TX_SIZE_SEARCH_METHOD); -typedef enum { +enum { // Try the full image with different values. LPF_PICK_FROM_FULL_IMAGE, + // Try the full image filter search with non-dual filter only. + LPF_PICK_FROM_FULL_IMAGE_NON_DUAL, // Try a small portion of the image with different values. LPF_PICK_FROM_SUBIMAGE, // Estimate the level based on quantizer and frame type LPF_PICK_FROM_Q, // Pick 0 to disable LPF if LPF was enabled last frame LPF_PICK_MINIMAL_LPF -} LPF_PICK_METHOD; +} UENUM1BYTE(LPF_PICK_METHOD); -typedef enum { +enum { + CDEF_FULL_SEARCH, + CDEF_FAST_SEARCH_LVL1, // Search among a subset of all possible filters. + CDEF_FAST_SEARCH_LVL2, // Search reduced subset of filters than Level 1. + CDEF_PICK_FROM_Q, // Estimate filter strength based on quantizer. + CDEF_PICK_METHODS +} UENUM1BYTE(CDEF_PICK_METHOD); + +enum { // Terminate search early based on distortion so far compared to // qp step, distortion in the neighborhood of the frame, etc. FLAG_EARLY_TERMINATE = 1 << 0, @@ -152,26 +179,28 @@ typedef enum { // Skips intra modes other than DC_PRED if the source variance is small FLAG_SKIP_INTRA_LOWVAR = 1 << 5, -} MODE_SEARCH_SKIP_LOGIC; +} UENUM1BYTE(MODE_SEARCH_SKIP_LOGIC); -typedef enum { +enum { NO_PRUNE = 0, - // eliminates one tx type in vertical and horizontal direction - PRUNE_ONE = 1, - // eliminates two tx types in each direction - PRUNE_TWO = 2, // adaptively prunes the least perspective tx types out of all 16 // (tuned to provide negligible quality loss) - PRUNE_2D_ACCURATE = 3, + PRUNE_2D_ACCURATE = 1, // similar, but applies much more aggressive pruning to get better speed-up - PRUNE_2D_FAST = 4, -} TX_TYPE_PRUNE_MODE; + PRUNE_2D_FAST = 2, + PRUNE_2D_MORE = 3, + // More aggressive pruning based on tx type score and allowed tx count + PRUNE_2D_AGGRESSIVE = 4, +} UENUM1BYTE(TX_TYPE_PRUNE_MODE); typedef struct { TX_TYPE_PRUNE_MODE prune_mode; int fast_intra_tx_type_search; int fast_inter_tx_type_search; + // prune two least frequently chosen transforms for each intra mode + int use_reduced_intra_txset; + // Use a skip flag prediction model to detect blocks with skip = 1 early // and avoid doing full TX type search for such blocks. int use_skip_flag_prediction; @@ -182,87 +211,67 @@ typedef struct { // skip remaining transform type search when we found the rdcost of skip is // better than applying transform int skip_tx_search; + + // Prune tx type search using previous frame stats. + int prune_tx_type_using_stats; + // Prune tx type search using estimated RDcost + int prune_tx_type_est_rd; + + // Flag used to control the winner mode processing for tx type pruning for + // inter blocks. It enables further tx type mode pruning based on ML model for + // mode evaluation and disables tx type mode pruning for winner mode + // processing. + int enable_winner_mode_tx_type_pruning; } TX_TYPE_SEARCH; -typedef enum { +enum { // Search partitions using RD criterion SEARCH_PARTITION, // Always use a fixed size partition FIXED_PARTITION, - REFERENCE_PARTITION -} PARTITION_SEARCH_TYPE; + REFERENCE_PARTITION, -typedef struct MV_SPEED_FEATURES { - // Motion search method (Diamond, NSTEP, Hex, Big Diamond, Square, etc). - SEARCH_METHODS search_method; - - // This parameter controls which step in the n-step process we start at. - // It's changed adaptively based on circumstances. - int reduce_first_step_size; - - // If this is set to 1, we limit the motion search range to 2 times the - // largest motion vector found in the last frame. - int auto_mv_step_size; - - // Subpel_search_method can only be subpel_tree which does a subpixel - // logarithmic search that keeps stepping at 1/2 pixel units until - // you stop getting a gain, and then goes on to 1/4 and repeats - // the same process. Along the way it skips many diagonals. - SUBPEL_SEARCH_METHODS subpel_search_method; + VAR_BASED_PARTITION +} UENUM1BYTE(PARTITION_SEARCH_TYPE); - // Maximum number of steps in logarithmic subpel search before giving up. - int subpel_iters_per_step; - - // Control when to stop subpel search - int subpel_force_stop; -} MV_SPEED_FEATURES; - -#define MAX_MESH_STEP 4 - -typedef struct MESH_PATTERN { - int range; - int interval; -} MESH_PATTERN; - -typedef enum { - GM_FULL_SEARCH, - GM_REDUCED_REF_SEARCH, - GM_DISABLE_SEARCH -} GM_SEARCH_TYPE; - -typedef enum { - GM_ERRORADV_TR_0, - GM_ERRORADV_TR_1, - GM_ERRORADV_TR_2, - GM_ERRORADV_TR_TYPES, -} GM_ERRORADV_TYPE; - -typedef enum { - NO_TRELLIS_OPT, // No trellis optimization - FULL_TRELLIS_OPT, // Trellis optimization in all stages - FINAL_PASS_TRELLIS_OPT // Trellis optimization in only the final encode pass -} TRELLIS_OPT_TYPE; - -typedef enum { - FULL_TXFM_RD, - LOW_TXFM_RD, -} TXFM_RD_MODEL; +enum { + NOT_IN_USE, + DIRECT_PRED, + RELAXED_PRED, + ADAPT_PRED +} UENUM1BYTE(MAX_PART_PRED_MODE); -typedef struct SPEED_FEATURES { - MV_SPEED_FEATURES mv; +enum { + LAST_MV_DATA, + CURRENT_Q, + QTR_ONLY, +} UENUM1BYTE(MV_PREC_LOGIC); +typedef struct HIGH_LEVEL_SPEED_FEATURES { // Frame level coding parameter update int frame_parameter_update; RECODE_LOOP_TYPE recode_loop; - // Trellis (dynamic programming) optimization of quantized values - TRELLIS_OPT_TYPE optimize_coefficients; + // This feature controls the tolerence vs target used in deciding whether to + // recode a frame. It has no meaning if recode is disabled. + int recode_tolerance; - // Global motion warp error threshold - GM_ERRORADV_TYPE gm_erroradv_type; + // Determine how motion vector precision is chosen. The possibilities are: + // LAST_MV_DATA: use the mv data from the last coded frame + // CURRENT_Q: use the current q as a threshold + // QTR_ONLY: use quarter pel precision only. + MV_PREC_LOGIC high_precision_mv_usage; + + // Whether to disable overlay frames for filtered Altref frames, + // overiding oxcf->enable_overlay flag set as 1. + int disable_overlay_frames; + + // Enable/disable adaptively deciding whether or not to encode ALTREF overlay + // frame. + int adaptive_overlay_encoding; // Always set to 0. If on it enables 0 cost background transmission // (except for the initial transmission of the segmentation). The feature is @@ -271,72 +280,58 @@ typedef struct SPEED_FEATURES { // adds overhead. int static_segmentation; - // Limit the inter mode tested in the RD loop - int reduce_inter_modes; + // Enable/disable second_alt_ref temporal filtering. + int second_alt_ref_filtering; +} HIGH_LEVEL_SPEED_FEATURES; + +typedef struct TPL_SPEED_FEATURES { + // Prune the intra modes search by tpl. + // If set to 0, we will search all intra modes from DC_PRED to PAETH_PRED. + // If set to 1, we only search DC_PRED, V_PRED, and H_PRED. + int prune_intra_modes; + // This parameter controls which step in the n-step process we start at. + int reduce_first_step_size; + // Skip motion estimation based on the precision of center MVs and the + // difference between center MVs. + // If set to 0, motion estimation is skipped for duplicate center MVs + // (default). If set to 1, motion estimation is skipped for duplicate + // full-pixel center MVs. If set to 2, motion estimation is skipped if the + // difference between center MVs is less than the threshold. + int skip_alike_starting_mv; + + // When to stop subpel search. + SUBPEL_FORCE_STOP subpel_force_stop; +} TPL_SPEED_FEATURES; + +typedef struct GLOBAL_MOTION_SPEED_FEATURES { + // Global motion warp error threshold + GM_ERRORADV_TYPE gm_erroradv_type; + + // Disable adaptive threshold for global motion warp error + int disable_adaptive_warp_error_thresh; // Do not compute the global motion parameters for a LAST2_FRAME or // LAST3_FRAME if the GOLDEN_FRAME is closer and it has a non identity // global model. int selective_ref_gm; - // If 1 we iterate finding a best reference for 2 ref frames together - via - // a log search that iterates 4 times (check around mv for last for best - // error of combined predictor then check around mv for alt). If 0 we - // we just use the best motion vector found for each frame by itself. - BLOCK_SIZE comp_inter_joint_search_thresh; - - // This variable is used to cap the maximum number of times we skip testing a - // mode to be evaluated. A high value means we will be faster. - int adaptive_rd_thresh; + GM_SEARCH_TYPE gm_search_type; - // Determine which method we use to determine transform size. We can choose - // between options like full rd, largest for prediction size, largest - // for intra and model coefs for the rest. - TX_SIZE_SEARCH_METHOD tx_size_search_method; + // whether to disable the global motion recode loop + int gm_disable_recode; - // Init search depth for square and rectangular transform partitions. - // Values: - // 0 - search full tree, 1: search 1 level, 2: search the highest level only - int inter_tx_size_search_init_depth_sqr; - int inter_tx_size_search_init_depth_rect; - int intra_tx_size_search_init_depth_sqr; - int intra_tx_size_search_init_depth_rect; - // If any dimension of a coding block size above 64, always search the - // largest transform only, since the largest transform block size is 64x64. - int tx_size_search_lgr_block; + // During global motion estimation, prune remaining reference frames in a + // given direction(past/future), if the evaluated ref_frame in that direction + // yields gm_type as INVALID/TRANSLATION/IDENTITY + int prune_ref_frame_for_gm_search; +} GLOBAL_MOTION_SPEED_FEATURES; +typedef struct PARTITION_SPEED_FEATURES { PARTITION_SEARCH_TYPE partition_search_type; - TX_TYPE_SEARCH tx_type_search; - - // Skip split transform block partition when the collocated bigger block - // is selected as all zero coefficients. - int txb_split_cap; - - // Shortcut the transform block partition and type search when the target - // rdcost is relatively lower. - // Values are 0 (not used) , or 1 - 2 with progressively increasing - // aggressiveness - int adaptive_txb_search_level; - - // Prune level for tx_size_type search for inter based on rd model - // 0: no pruning - // 1-2: progressively increasing aggressiveness of pruning - int model_based_prune_tx_search_level; - - // Model based breakout after interpolation filter search - // 0: no breakout - // 1: use model based rd breakout - int model_based_post_interp_filter_breakout; - // Used if partition_search_type = FIXED_SIZE_PARTITION BLOCK_SIZE always_this_block_size; - // Drop less likely to be picked reference frames in the RD search. - // Has three levels for now: 0, 1 and 2, where higher levels prune more - // aggressively than lower ones. (0 means no pruning). - int selective_ref_frame; - // Prune extended partition types search // Can take values 0 - 2, 0 referring to no pruning, and 1 - 2 increasing // aggressiveness of pruning in order. @@ -351,14 +346,10 @@ typedef struct SPEED_FEATURES { // Use a ML model to prune horz4 and vert4 partitions. int ml_prune_4_partition; - int fast_cdef_search; - - // 2-pass coding block partition search - int two_pass_partition_search; - - // Use the mode decisions made in the initial partition search to prune mode - // candidates, e.g. ref frames. - int mode_pruning_based_on_two_pass_partition_search; + // Use a ML model to adaptively terminate partition search after trying + // PARTITION_SPLIT. Can take values 0 - 2, 0 meaning not being enabled, and + // 1 - 2 increasing aggressiveness in order. + int ml_early_term_after_part_split_level; // Skip rectangular partition test when partition type none gives better // rd than partition type split. Can take values 0 - 2, 0 referring to no @@ -368,145 +359,408 @@ typedef struct SPEED_FEATURES { // Use square partition only beyond this block size. BLOCK_SIZE use_square_partition_only_threshold; - // Prune reference frames for rectangular partitions. - int prune_ref_frame_for_rect_partitions; - - // Sets min and max partition sizes for this superblock based on the - // same superblock in last encoded frame, and the left and above neighbor. - AUTO_MIN_MAX_MODE auto_min_max_partition_size; - // Ensures the rd based auto partition search will always - // go down at least to the specified level. - BLOCK_SIZE rd_auto_partition_min_limit; + // Sets min and max square partition levels for this superblock based on + // motion vector and prediction error distribution produced from 16x16 + // simple motion search + MAX_PART_PRED_MODE auto_max_partition_based_on_simple_motion; + int auto_min_partition_based_on_simple_motion; - // Min and max partition size we enable (block_size) as per auto + // Min and max square partition size we enable (block_size) as per auto // min max, but also used by adjust partitioning, and pick_partitioning. BLOCK_SIZE default_min_partition_size; BLOCK_SIZE default_max_partition_size; - // Whether or not we allow partitions one smaller or one greater than the last - // frame's partitioning. Only used if use_lastframe_partitioning is set. - int adjust_partitioning_from_last_frame; + // Sets level of adjustmet of variace-based partitioning during + // rd_use_partition 0 - no partition adjusment, 1 - try to merge partitions + // for small blocks and high QP, 2 - always try to merge leaf partitions, 3 - + // try to merge and split leaf partitions + int adjust_var_based_rd_partitioning; + + // Partition search early breakout thresholds. + int64_t partition_search_breakout_dist_thr; + int partition_search_breakout_rate_thr; + + // Thresholds for ML based partition search breakout. + int ml_partition_search_breakout_thresh[PARTITION_BLOCK_SIZES]; + + // Allow skipping partition search for still image frame + int allow_partition_search_skip; + + // The aggresiveness of pruning with simple_motion_search. + // Currently 0 is the lowest, and 2 the highest. + int simple_motion_search_prune_agg; - // Disables sub 8x8 blocksizes in different scenarios: Choices are to disable - // it always, to allow it for only Last frame and Intra, disable it for all - // inter modes or to enable it always. - int disable_split_mask; + // Perform simple_motion_search on each possible subblock and use it to prune + // PARTITION_HORZ and PARTITION_VERT. + int simple_motion_search_prune_rect; + + // Perform simple motion search before none_partition to decide if we + // want to remove all partitions other than PARTITION_SPLIT. If set to 0, this + // model is disabled. If set to 1, the model attempts to perform + // PARTITION_SPLIT only. If set to 2, the model also attempts to prune + // PARTITION_SPLIT. + int simple_motion_search_split; + + // Use features from simple_motion_search to terminate prediction block + // partition after PARTITION_NONE + int simple_motion_search_early_term_none; + + // This variable controls the maximum block size where intra blocks can be + // used in inter frames. + // TODO(aconverse): Fold this into one of the other many mode skips + BLOCK_SIZE max_intra_bsize; + + // Use CNN with luma pixels on source frame on each of the 64x64 subblock to + // perform split/no_split decision on intra-frames. + int intra_cnn_split; + + // Disable extended partition search for lower block sizes. + int ext_partition_eval_thresh; + + // Prune 1:4 partition search based on winner info from split partitions + int prune_4_partition_using_split_info; + + // Prune AB partition search using split and HORZ/VERT info + int prune_ab_partition_using_split_info; +} PARTITION_SPEED_FEATURES; + +typedef struct MV_SPEED_FEATURES { + // Motion search method (Diamond, NSTEP, Hex, Big Diamond, Square, etc). + SEARCH_METHODS search_method; + + // If this is set to 1, we limit the motion search range to 2 times the + // largest motion vector found in the last frame. + int auto_mv_step_size; + + // Subpel_search_method can only be subpel_tree which does a subpixel + // logarithmic search that keeps stepping at 1/2 pixel units until + // you stop getting a gain, and then goes on to 1/4 and repeats + // the same process. Along the way it skips many diagonals. + SUBPEL_SEARCH_METHODS subpel_search_method; + + // Maximum number of steps in logarithmic subpel search before giving up. + int subpel_iters_per_step; + + // When to stop subpel search. + SUBPEL_FORCE_STOP subpel_force_stop; + + // When to stop subpel search in simple motion search. + SUBPEL_FORCE_STOP simple_motion_subpel_force_stop; + + // If true, sub-pixel search uses the exact convolve function used for final + // encoding and decoding; otherwise, it uses bilinear interpolation. + SUBPEL_SEARCH_TYPE use_accurate_subpel_search; // TODO(jingning): combine the related motion search speed features // This allows us to use motion search at other sizes as a starting // point for this motion search and limits the search range around it. int adaptive_motion_search; - // Flag for allowing some use of exhaustive searches; - int allow_exhaustive_searches; - - // Threshold for allowing exhaistive motion search. + // Threshold for allowing exhaustive motion search. int exhaustive_searches_thresh; - // Maximum number of exhaustive searches for a frame. - int max_exaustive_pct; - - // Pattern to be used for any exhaustive mesh searches. + // Pattern to be used for any exhaustive mesh searches (except intraBC ME). MESH_PATTERN mesh_patterns[MAX_MESH_STEP]; - // Allows sub 8x8 modes to use the prediction filter that was determined - // best for 8x8 mode. If set to 0 we always re check all the filters for - // sizes less than 8x8, 1 means we check all filter modes if no 8x8 filter - // was selected, and 2 means we use 8 tap if no 8x8 filter mode was selected. - int adaptive_pred_interp_filter; + // Pattern to be used for exhaustive mesh searches of intraBC ME. + MESH_PATTERN intrabc_mesh_patterns[MAX_MESH_STEP]; + + // Reduce single motion search range based on MV result of prior ref_mv_idx. + int reduce_search_range; + + // Prune mesh search. + int prune_mesh_search; + + // Use the rd cost around the best FULLPEL_MV to speed up subpel search + int use_fullpel_costlist; + + // Set the full pixel search level of obmc + // 0: obmc_full_pixel_diamond + // 1: obmc_refining_search_sad (faster) + int obmc_full_pixel_search_level; + + // Accurate full pixel motion search based on TPL stats. + int full_pixel_search_level; +} MV_SPEED_FEATURES; + +typedef struct INTER_MODE_SPEED_FEATURES { + // 2-pass inter mode model estimation where the preliminary pass skips + // transform search and uses a model to estimate rd, while the final pass + // computes the full transform search. Two types of models are supported: + // 0: not used + // 1: used with online dynamic rd model + // 2: used with static rd model + int inter_mode_rd_model_estimation; + + // Bypass transform search based on skip rd + int txfm_rd_gate_level; + + // Limit the inter mode tested in the RD loop + int reduce_inter_modes; // Adaptive prediction mode search int adaptive_mode_search; - int cb_partition_search; + // This variable is used to cap the maximum number of times we skip testing a + // mode to be evaluated. A high value means we will be faster. + int adaptive_rd_thresh; + + // Aggressively prune inter modes when best mode is skippable. + int prune_inter_modes_if_skippable; + + // Drop less likely to be picked reference frames in the RD search. + // Has five levels for now: 0, 1, 2, 3 and 4, where higher levels prune more + // aggressively than lower ones. (0 means no pruning). + int selective_ref_frame; + + // Prune reference frames for rectangular partitions. + // 0 implies no pruning + // 1 implies prune for extended partition + // 2 implies prune horiz, vert and extended partition + int prune_ref_frame_for_rect_partitions; int alt_ref_search_fp; - // Implements various heuristics to skip searching modes - // The heuristics selected are based on flags - // defined in the MODE_SEARCH_SKIP_HEURISTICS enum - unsigned int mode_search_skip_flags; + // flag to skip NEWMV mode in drl if the motion search result is the same + int skip_repeated_newmv; - // A source variance threshold below which filter search is disabled - // Choose a very large value (UINT_MAX) to use 8-tap always - unsigned int disable_filter_search_var_thresh; + // Skip the current ref_mv in NEW_MV mode if we have already encountered + // another ref_mv in the drl such that: + // 1. The other drl has the same fullpel_mv during the SIMPLE_TRANSLATION + // search process as the current fullpel_mv. + // 2. The rate needed to encode the current fullpel_mv is larger than that + // for the other ref_mv. + int skip_repeated_full_newmv; + + // This speed feature checks duplicate ref MVs among NEARESTMV, NEARMV, + // GLOBALMV and skips NEARMV or GLOBALMV (in order) if a duplicate is found + // TODO(any): Instead of skipping repeated ref mv, use the recalculated + // rd-cost based on mode rate and skip the mode evaluation + int skip_repeated_ref_mv; + + // Flag used to control the ref_best_rd based gating for chroma + int perform_best_rd_based_gating_for_chroma; + + // Skip certain motion modes (OBMC, warped, interintra) for single reference + // motion search, using the results of single ref SIMPLE_TRANSLATION + int prune_single_motion_modes_by_simple_trans; + + // Reuse the inter_intra_mode search result from NEARESTMV mode to other + // single ref modes + int reuse_inter_intra_mode; + + // prune wedge and compound segment approximate rd evaluation based on + // compound average modeled rd + int prune_comp_type_by_model_rd; + + // prune wedge and compound segment approximate rd evaluation based on + // compound average rd/ref_best_rd + int prune_comp_type_by_comp_avg; - // A source variance threshold below which wedge search is disabled + // Skip some ref frames in compound motion search by single motion search + // result. Has three levels for now: 0 referring to no skipping, and 1 - 3 + // increasing aggressiveness of skipping in order. + // Note: The search order might affect the result. It assumes that the single + // reference modes are searched before compound modes. It is better to search + // same single inter mode as a group. + int prune_comp_search_by_single_result; + + // If 1 we iterate finding a best reference for 2 ref frames together - via + // a log search that iterates 4 times (check around mv for last for best + // error of combined predictor then check around mv for alt). If 0 we + // we just use the best motion vector found for each frame by itself. + BLOCK_SIZE comp_inter_joint_search_thresh; + + // Instead of performing a full MV search, do a simple translation first + // and only perform a full MV search on the motion vectors that performed + // well. + int prune_mode_search_simple_translation; + + // Only search compound modes with at least one "good" reference frame. + // A reference frame is good if, after looking at its performance among + // the single reference modes, it is one of the two best performers. + int prune_compound_using_single_ref; + + // Skip extended compound mode using ref frames of above and left neighbor + // blocks. + // 0 : no pruning + // 1 : prune extended compound mode (less aggressiveness) + // 2 : prune extended compound mode (high aggressiveness) + int prune_compound_using_neighbors; + + // Based on previous ref_mv_idx search result, prune the following search. + int prune_ref_mv_idx_search; + + // Disable one sided compound modes. + int disable_onesided_comp; + + // Prune/gate motion mode evaluation based on token based rd + // during transform search for inter blocks + // Values are 0 (not used) , 1 - 3 with progressively increasing + // aggressiveness + int prune_motion_mode_level; + + // Prune obmc search using previous frame stats. + int prune_obmc_prob_thresh; + + // Disable obmc. + int disable_obmc; + + // Gate warp evaluation for motions of type IDENTITY, + // TRANSLATION and AFFINE(based on number of warp neighbors) + int prune_warp_using_wmtype; + + // Prune warped motion search using previous frame stats. + int prune_warped_prob_thresh; + + // Enable/disable interintra wedge search. + int disable_wedge_interintra_search; + + // De-couple wedge and mode search during interintra RDO. + int fast_interintra_wedge_search; + + // Only enable wedge search if the edge strength is greater than + // this threshold. A value of 0 signals that this check is disabled. + unsigned int disable_wedge_search_edge_thresh; + + // Only enable wedge search if the variance is above this threshold. unsigned int disable_wedge_search_var_thresh; // Whether fast wedge sign estimate is used int fast_wedge_sign_estimate; - // These bit masks allow you to enable or disable intra modes for each - // transform size separately. - int intra_y_mode_mask[TX_SIZES]; - int intra_uv_mode_mask[TX_SIZES]; + // Whether to prune wedge search based on predictor difference + int prune_wedge_pred_diff_based; - // This feature controls how the loop filter level is determined. - LPF_PICK_METHOD lpf_pick; + // Enable/disable ME for interinter wedge search. + int disable_interinter_wedge_newmv_search; - // This feature controls whether we do the expensive context update and - // calculation in the rd coefficient costing loop. - int use_fast_coef_costing; + // Enable/disable ME for interinter diffwtd search. PSNR BD-rate gain of + // ~0.1 on the lowres test set, but ~15% slower computation. + int enable_interinter_diffwtd_newmv_search; - // This feature controls the tolerence vs target used in deciding whether to - // recode a frame. It has no meaning if recode is disabled. - int recode_tolerance; + // Enable/disable smooth inter-intra mode + int disable_smooth_interintra; - // This variable controls the maximum block size where intra blocks can be - // used in inter frames. - // TODO(aconverse): Fold this into one of the other many mode skips - BLOCK_SIZE max_intra_bsize; + // Disable interinter_wedge + int disable_interinter_wedge; - // Partition search early breakout thresholds. - int64_t partition_search_breakout_dist_thr; - int partition_search_breakout_rate_thr; + // Decide when and how to use joint_comp. + DIST_WTD_COMP_FLAG use_dist_wtd_comp_flag; - // Thresholds for ML based partition search breakout. - int ml_partition_search_breakout_thresh[PARTITION_BLOCK_SIZES]; + // Whether to override and disable sb level coeff cost updates, if + // cpi->oxcf.coeff_cost_upd_freq = COST_UPD_SB (i.e. set at SB level) + int disable_sb_level_coeff_cost_upd; - // Allow skipping partition search for still image frame - int allow_partition_search_skip; + // Whether to override and disable sb level mv cost updates, if + // cpi->oxcf.coeff_cost_upd_freq = COST_UPD_SB (i.e. set at SB level) + int disable_sb_level_mv_cost_upd; - // Fast approximation of av1_model_rd_from_var_lapndz - int simple_model_rd_from_var; - - // If true, sub-pixel search uses the exact convolve function used for final - // encoding and decoding; otherwise, it uses bilinear interpolation. - int use_accurate_subpel_search; + // Prune inter modes based on tpl stats + // 0 : no pruning + // 1 - 3 indicate increasing aggressiveness in order. + int prune_inter_modes_based_on_tpl; - // Whether to compute distortion in the image domain (slower but - // more accurate), or in the transform domain (faster but less acurate). - // 0: use image domain - // 1: use transform domain in tx_type search, and use image domain for - // RD_STATS - // 2: use transform domain - int use_transform_domain_distortion; + // Model based breakout after interpolation filter search + // 0: no breakout + // 1: use model based rd breakout + int model_based_post_interp_filter_breakout; - GM_SEARCH_TYPE gm_search_type; + // Reuse compound type rd decision when exact match is found + // 0: No reuse + // 1: Reuse the compound type decision + int reuse_compound_type_decision; +} INTER_MODE_SPEED_FEATURES; - // whether to disable the global motion recode loop - int gm_disable_recode; +typedef struct INTERP_FILTER_SPEED_FEATURES { + // A source variance threshold below which filter search is disabled + // Choose a very large value (UINT_MAX) to use 8-tap always + unsigned int disable_filter_search_var_thresh; // Do limited interpolation filter search for dual filters, since best choice // usually includes EIGHTTAP_REGULAR. int use_fast_interpolation_filter_search; - // Save results of interpolation_filter_search for a block - // Check mv and ref_frames before search, if they are same with previous - // saved results, it can be skipped. - int skip_repeat_interpolation_filter_search; + // Disable dual filter + int disable_dual_filter; - // Use a hash table to store previously computed optimized qcoeffs from - // expensive calls to optimize_txb. - int use_hash_based_trellis; + // Save results of av1_interpolation_filter_search for a block + // Check mv and ref_frames before search, if they are very close with previous + // saved results, filter search can be skipped. + int use_interp_filter; + + // skip sharp_filter evaluation based on regular and smooth filter rd for + // dual_filter=0 case + int skip_sharp_interp_filter_search; + + int cb_pred_filter_search; - // flag to drop some ref frames in compound motion search - int drop_ref; + // adaptive interp_filter search to allow skip of certain filter types. + int adaptive_interp_filter_search; +} INTERP_FILTER_SPEED_FEATURES; + +typedef struct INTRA_MODE_SPEED_FEATURES { + // These bit masks allow you to enable or disable intra modes for each + // transform size separately. + int intra_y_mode_mask[TX_SIZES]; + int intra_uv_mode_mask[TX_SIZES]; // flag to allow skipping intra mode for inter frame prediction int skip_intra_in_interframe; + // variance threshold for intra mode gating when inter turned out to be skip + // in inter frame prediction + unsigned int src_var_thresh_intra_skip; + + // Prune intra mode candidates based on source block histogram of gradient. + int intra_pruning_with_hog; + + // TODO(anyone): tune intra_pruning_with_hog_thresh for various speeds. + float intra_pruning_with_hog_thresh; + + // Enable/disable smooth intra modes. + int disable_smooth_intra; + + // prune palette search + // 0: No pruning + // 1: Perform coarse search to prune the palette colors. For winner colors, + // neighbors are also evaluated using a finer search. + // 2: Perform 2 way palette search from max colors to min colors (and min + // colors to remaining colors) and terminate the search if current number of + // palette colors is not the winner. + int prune_palette_search_level; +} INTRA_MODE_SPEED_FEATURES; + +typedef struct TX_SPEED_FEATURES { + // Init search depth for square and rectangular transform partitions. + // Values: + // 0 - search full tree, 1: search 1 level, 2: search the highest level only + int inter_tx_size_search_init_depth_sqr; + int inter_tx_size_search_init_depth_rect; + int intra_tx_size_search_init_depth_sqr; + int intra_tx_size_search_init_depth_rect; + + // If any dimension of a coding block size above 64, always search the + // largest transform only, since the largest transform block size is 64x64. + int tx_size_search_lgr_block; + + TX_TYPE_SEARCH tx_type_search; + + // Skip split transform block partition when the collocated bigger block + // is selected as all zero coefficients. + int txb_split_cap; + + // Shortcut the transform block partition and type search when the target + // rdcost is relatively lower. + // Values are 0 (not used) , or 1 - 2 with progressively increasing + // aggressiveness + int adaptive_txb_search_level; + + // Prune level for tx_size_type search for inter based on rd model + // 0: no pruning + // 1-2: progressively increasing aggressiveness of pruning + int model_based_prune_tx_search_level; + // Use hash table to store intra(keyframe only) txb transform search results // to avoid repeated search on the same residue signal. int use_intra_txb_hash; @@ -515,51 +769,263 @@ typedef struct SPEED_FEATURES { // to avoid repeated search on the same residue signal. int use_inter_txb_hash; + // Refine TX type after fast TX search. + int refine_fast_tx_search_results; +} TX_SPEED_FEATURES; + +typedef struct RD_CALC_SPEED_FEATURES { + // This feature controls whether we do the expensive context update and + // calculation in the rd coefficient costing loop. + int use_fast_coef_costing; + + // Fast approximation of av1_model_rd_from_var_lapndz + int simple_model_rd_from_var; + + // Whether to compute distortion in the image domain (slower but + // more accurate), or in the transform domain (faster but less acurate). + // 0: use image domain + // 1: use transform domain in tx_type search, and use image domain for + // RD_STATS + // 2: use transform domain + int tx_domain_dist_level; + + // Transform domain distortion threshold level + int tx_domain_dist_thres_level; + + // Trellis (dynamic programming) optimization of quantized values + TRELLIS_OPT_TYPE optimize_coefficients; + + // Use a hash table to store previously computed optimized qcoeffs from + // expensive calls to optimize_txb. + int use_hash_based_trellis; + // Use hash table to store macroblock RD search results // to avoid repeated search on the same residue signal. int use_mb_rd_hash; + // Flag used to control the speed of the eob selection in trellis. + int trellis_eob_fast; + // Calculate RD cost before doing optimize_b, and skip if the cost is large. int optimize_b_precheck; - // Use model rd instead of transform search in jnt_comp - int jnt_comp_fast_tx_search; + // Flag used to control the extent of coeff R-D optimization + int perform_coeff_opt; +} RD_CALC_SPEED_FEATURES; + +typedef struct WINNER_MODE_SPEED_FEATURES { + // Flag used to control the winner mode processing for better R-D optimization + // of quantized coeffs + int enable_winner_mode_for_coeff_opt; + + // Flag used to control the winner mode processing for transform size + // search method + int enable_winner_mode_for_tx_size_srch; + + // Control transform size search level + // Eval type: Default Mode Winner + // Level 0 : FULL RD LARGEST ALL FULL RD + // Level 1 : FAST RD LARGEST ALL FULL RD + // Level 2 : LARGEST ALL LARGEST ALL FULL RD + int tx_size_search_level; + + // Flag used to control the winner mode processing for use transform + // domain distortion + int enable_winner_mode_for_use_tx_domain_dist; - // Skip mv search in jnt_comp - int jnt_comp_skip_mv_search; + // Flag used to enable processing of multiple winner modes + int enable_multiwinner_mode_process; + + // Motion mode for winner candidates: + // 0: speed feature OFF + // 1 / 2 : Use configured number of winner candidates + int motion_mode_for_winner_cand; +} WINNER_MODE_SPEED_FEATURES; + +typedef struct LOOP_FILTER_SPEED_FEATURES { + // This feature controls how the loop filter level is determined. + LPF_PICK_METHOD lpf_pick; + + // Control how the CDEF strength is determined. + CDEF_PICK_METHOD cdef_pick_method; // Decoder side speed feature to add penalty for use of dual-sgr filters. // Takes values 0 - 10, 0 indicating no penalty and each additional level // adding a penalty of 1% int dual_sgr_penalty_level; - // Dynamically estimate final rd from prediction error and mode cost - int inter_mode_rd_model_estimation; + // prune sgr ep using binary search like mechanism + int enable_sgr_ep_pruning; - // Skip some ref frames in compound motion search by single motion search - // result. Has three levels for now: 0 referring to no skipping, and 1 - 3 - // increasing aggressiveness of skipping in order. - // Note: The search order might affect the result. It is better to search same - // single inter mode as a group. - int prune_comp_search_by_single_result; + // Disable loop restoration for Chroma plane + int disable_loop_restoration_chroma; - // Reuse the inter_intra_mode search result from NEARESTMV mode to other - // single ref modes - int reuse_inter_intra_mode; + // Prune RESTORE_WIENER evaluation based on source variance + // 0 : no pruning + // 1 : conservative pruning + // 2 : aggressive pruning + int prune_wiener_based_on_src_var; - // Set the full pixel search level of obmc - // 0: obmc_full_pixel_diamond - // 1: obmc_refining_search_sad (faster) - int obmc_full_pixel_search_level; + // Prune self-guided loop restoration based on wiener search results + // 0 : no pruning + // 1 : pruning based on rdcost ratio of RESTORE_WIENER and RESTORE_NONE + // 2 : pruning based on winner restoration type among RESTORE_WIENER and + // RESTORE_NONE + int prune_sgr_based_on_wiener; - // flag to skip NEWMV mode in drl if the motion search result is the same - int skip_repeated_newmv; + // Reduce the wiener filter win size for luma + int reduce_wiener_window_size; + + // Disable loop restoration filter + int disable_lr_filter; +} LOOP_FILTER_SPEED_FEATURES; + +typedef struct REAL_TIME_SPEED_FEATURES { + // check intra prediction for non-RD mode. + int check_intra_pred_nonrd; + + // Perform coarse ME before calculating variance in variance-based partition + int estimate_motion_for_var_based_partition; + + // For nonrd_use_partition: mode of extra check of leaf partition + // 0 - don't check merge + // 1 - always check merge + // 2 - check merge and prune checking final split + int nonrd_check_partition_merge_mode; + + // For nonrd_use_partition: check of leaf partition extra split + int nonrd_check_partition_split; + + // Implements various heuristics to skip searching modes + // The heuristics selected are based on flags + // defined in the MODE_SEARCH_SKIP_HEURISTICS enum + unsigned int mode_search_skip_flags; + + // For nonrd: Reduces ref frame search. + // 0 - low level of search prune in non last frames + // 1 - pruned search in non last frames + // 2 - more pruned search in non last frames + int nonrd_prune_ref_frame_search; + + // This flag controls the use of non-RD mode decision. + int use_nonrd_pick_mode; + + // Use ALTREF frame in non-RD mode decision. + int use_nonrd_altref_frame; + + // Use compound reference for non-RD mode. + int use_comp_ref_nonrd; + + // use reduced ref set for real-time mode + int use_real_time_ref_set; + + // Skip a number of expensive mode evaluations for blocks with very low + // temporal variance. + int short_circuit_low_temp_var; + + // Use modeled (currently CurvFit model) RDCost for fast non-RD mode + int use_modeled_non_rd_cost; + + // Reuse inter prediction in fast non-rd mode. + int reuse_inter_pred_nonrd; + + // Number of best inter modes to search transform. INT_MAX - search all. + int num_inter_modes_for_tx_search; + + // Forces TX search off for RDCost calulation. + int force_tx_search_off; + + // Use interpolation filter search in non-RD mode decision. + int use_nonrd_filter_search; + + // Use simplified RD model for interpolation search and Intra + int use_simple_rd_model; + + // If set forces interpolation filter to EIGHTTAP_REGULAR + int skip_interp_filter_search; + + // Use hybrid (rd for bsize < 16x16, otherwise nonrd) intra search for intra + // only frames. + int hybrid_intra_pickmode; + + // Compute variance/sse on source difference, prior to encoding superblock. + int source_metrics_sb_nonrd; +} REAL_TIME_SPEED_FEATURES; + +typedef struct SPEED_FEATURES { + /* + * Sequence/frame level speed features: + */ + HIGH_LEVEL_SPEED_FEATURES hl_sf; + + /* + * Speed features related to how tpl's searches are done. + */ + TPL_SPEED_FEATURES tpl_sf; + + /* + * Global motion speed features: + */ + GLOBAL_MOTION_SPEED_FEATURES gm_sf; + + /* + * Partition search speed features: + */ + PARTITION_SPEED_FEATURES part_sf; + + /* + * Motion search speed features: + */ + MV_SPEED_FEATURES mv_sf; + + /* + * Inter mode search speed features: + */ + INTER_MODE_SPEED_FEATURES inter_sf; + + /* + * Interpolation filter search speed features: + */ + INTERP_FILTER_SPEED_FEATURES interp_sf; + + /* + * Intra mode search speed features: + */ + INTRA_MODE_SPEED_FEATURES intra_sf; + + /* + * Transform size/type search speed features: + */ + TX_SPEED_FEATURES tx_sf; + + /* + * RD calculation speed features: + */ + RD_CALC_SPEED_FEATURES rd_sf; + + /* + * Two-pass mode evaluation features: + */ + WINNER_MODE_SPEED_FEATURES winner_mode_sf; + + /* + * In-loop filter speed features: + */ + LOOP_FILTER_SPEED_FEATURES lpf_sf; + + /* + * Real-time mode speed features: + */ + REAL_TIME_SPEED_FEATURES rt_sf; } SPEED_FEATURES; struct AV1_COMP; -void av1_set_speed_features_framesize_independent(struct AV1_COMP *cpi); -void av1_set_speed_features_framesize_dependent(struct AV1_COMP *cpi); +void av1_set_speed_features_framesize_independent(struct AV1_COMP *cpi, + int speed); +void av1_set_speed_features_framesize_dependent(struct AV1_COMP *cpi, + int speed); +void av1_set_speed_features_qindex_dependent(struct AV1_COMP *cpi, int speed); #ifdef __cplusplus } // extern "C" |