Remove counts param
[aom.git] / vpx / vpx_encoder.h
blobbf75584d58919b407d4758ec650c5a2a0fc84bed
1 /*
2 * Copyright (c) 2010 The WebM project authors. All Rights Reserved.
4 * Use of this source code is governed by a BSD-style license
5 * that can be found in the LICENSE file in the root of the source
6 * tree. An additional intellectual property rights grant can be found
7 * in the file PATENTS. All contributing project authors may
8 * be found in the AUTHORS file in the root of the source tree.
9 */
10 #ifndef VPX_VPX_ENCODER_H_
11 #define VPX_VPX_ENCODER_H_
13 /*!\defgroup encoder Encoder Algorithm Interface
14 * \ingroup codec
15 * This abstraction allows applications using this encoder to easily support
16 * multiple video formats with minimal code duplication. This section describes
17 * the interface common to all encoders.
18 * @{
21 /*!\file
22 * \brief Describes the encoder algorithm interface to applications.
24 * This file describes the interface between an application and a
25 * video encoder algorithm.
28 #ifdef __cplusplus
29 extern "C" {
30 #endif
32 #include "./vpx_codec.h"
34 /*! Temporal Scalability: Maximum length of the sequence defining frame
35 * layer membership
37 #define VPX_TS_MAX_PERIODICITY 16
39 /*! Temporal Scalability: Maximum number of coding layers */
40 #define VPX_TS_MAX_LAYERS 5
42 /*!\deprecated Use #VPX_TS_MAX_PERIODICITY instead. */
43 #define MAX_PERIODICITY VPX_TS_MAX_PERIODICITY
45 /*!\deprecated Use #VPX_TS_MAX_LAYERS instead. */
46 #define MAX_LAYERS VPX_TS_MAX_LAYERS
48 /*! Spatial Scalability: Maximum number of coding layers */
49 #define VPX_SS_MAX_LAYERS 5
51 /*! Spatial Scalability: Default number of coding layers */
52 #define VPX_SS_DEFAULT_LAYERS 1
54 /*!\brief Current ABI version number
56 * \internal
57 * If this file is altered in any way that changes the ABI, this value
58 * must be bumped. Examples include, but are not limited to, changing
59 * types, removing or reassigning enums, adding/removing/rearranging
60 * fields to structures
62 #define VPX_ENCODER_ABI_VERSION (4 + VPX_CODEC_ABI_VERSION) /**<\hideinitializer*/
65 /*! \brief Encoder capabilities bitfield
67 * Each encoder advertises the capabilities it supports as part of its
68 * ::vpx_codec_iface_t interface structure. Capabilities are extra
69 * interfaces or functionality, and are not required to be supported
70 * by an encoder.
72 * The available flags are specified by VPX_CODEC_CAP_* defines.
74 #define VPX_CODEC_CAP_PSNR 0x10000 /**< Can issue PSNR packets */
76 /*! Can output one partition at a time. Each partition is returned in its
77 * own VPX_CODEC_CX_FRAME_PKT, with the FRAME_IS_FRAGMENT flag set for
78 * every partition but the last. In this mode all frames are always
79 * returned partition by partition.
81 #define VPX_CODEC_CAP_OUTPUT_PARTITION 0x20000
83 /*! Can support input images at greater than 8 bitdepth.
85 #define VPX_CODEC_CAP_HIGHBITDEPTH 0x40000
87 /*! \brief Initialization-time Feature Enabling
89 * Certain codec features must be known at initialization time, to allow
90 * for proper memory allocation.
92 * The available flags are specified by VPX_CODEC_USE_* defines.
94 #define VPX_CODEC_USE_PSNR 0x10000 /**< Calculate PSNR on each frame */
95 #define VPX_CODEC_USE_OUTPUT_PARTITION 0x20000 /**< Make the encoder output one
96 partition at a time. */
97 #define VPX_CODEC_USE_HIGHBITDEPTH 0x40000 /**< Use high bitdepth */
100 /*!\brief Generic fixed size buffer structure
102 * This structure is able to hold a reference to any fixed size buffer.
104 typedef struct vpx_fixed_buf {
105 void *buf; /**< Pointer to the data */
106 size_t sz; /**< Length of the buffer, in chars */
107 } vpx_fixed_buf_t; /**< alias for struct vpx_fixed_buf */
110 /*!\brief Time Stamp Type
112 * An integer, which when multiplied by the stream's time base, provides
113 * the absolute time of a sample.
115 typedef int64_t vpx_codec_pts_t;
118 /*!\brief Compressed Frame Flags
120 * This type represents a bitfield containing information about a compressed
121 * frame that may be useful to an application. The most significant 16 bits
122 * can be used by an algorithm to provide additional detail, for example to
123 * support frame types that are codec specific (MPEG-1 D-frames for example)
125 typedef uint32_t vpx_codec_frame_flags_t;
126 #define VPX_FRAME_IS_KEY 0x1 /**< frame is the start of a GOP */
127 #define VPX_FRAME_IS_DROPPABLE 0x2 /**< frame can be dropped without affecting
128 the stream (no future frame depends on
129 this one) */
130 #define VPX_FRAME_IS_INVISIBLE 0x4 /**< frame should be decoded but will not
131 be shown */
132 #define VPX_FRAME_IS_FRAGMENT 0x8 /**< this is a fragment of the encoded
133 frame */
135 /*!\brief Error Resilient flags
137 * These flags define which error resilient features to enable in the
138 * encoder. The flags are specified through the
139 * vpx_codec_enc_cfg::g_error_resilient variable.
141 typedef uint32_t vpx_codec_er_flags_t;
142 #define VPX_ERROR_RESILIENT_DEFAULT 0x1 /**< Improve resiliency against
143 losses of whole frames */
144 #define VPX_ERROR_RESILIENT_PARTITIONS 0x2 /**< The frame partitions are
145 independently decodable by the
146 bool decoder, meaning that
147 partitions can be decoded even
148 though earlier partitions have
149 been lost. Note that intra
150 predicition is still done over
151 the partition boundary. */
153 /*!\brief Encoder output packet variants
155 * This enumeration lists the different kinds of data packets that can be
156 * returned by calls to vpx_codec_get_cx_data(). Algorithms \ref MAY
157 * extend this list to provide additional functionality.
159 enum vpx_codec_cx_pkt_kind {
160 VPX_CODEC_CX_FRAME_PKT, /**< Compressed video frame */
161 VPX_CODEC_STATS_PKT, /**< Two-pass statistics for this frame */
162 VPX_CODEC_FPMB_STATS_PKT, /**< first pass mb statistics for this frame */
163 VPX_CODEC_PSNR_PKT, /**< PSNR statistics for this frame */
164 // Spatial SVC is still experimental and may be removed before the next ABI
165 // bump.
166 #if VPX_ENCODER_ABI_VERSION > (4 + VPX_CODEC_ABI_VERSION)
167 VPX_CODEC_SPATIAL_SVC_LAYER_SIZES, /**< Sizes for each layer in this frame*/
168 VPX_CODEC_SPATIAL_SVC_LAYER_PSNR, /**< PSNR for each layer in this frame*/
169 #endif
170 VPX_CODEC_CUSTOM_PKT = 256 /**< Algorithm extensions */
174 /*!\brief Encoder output packet
176 * This structure contains the different kinds of output data the encoder
177 * may produce while compressing a frame.
179 typedef struct vpx_codec_cx_pkt {
180 enum vpx_codec_cx_pkt_kind kind; /**< packet variant */
181 union {
182 struct {
183 void *buf; /**< compressed data buffer */
184 size_t sz; /**< length of compressed data */
185 vpx_codec_pts_t pts; /**< time stamp to show frame
186 (in timebase units) */
187 unsigned long duration; /**< duration to show frame
188 (in timebase units) */
189 vpx_codec_frame_flags_t flags; /**< flags for this frame */
190 int partition_id; /**< the partition id
191 defines the decoding order
192 of the partitions. Only
193 applicable when "output partition"
194 mode is enabled. First partition
195 has id 0.*/
197 } frame; /**< data for compressed frame packet */
198 vpx_fixed_buf_t twopass_stats; /**< data for two-pass packet */
199 vpx_fixed_buf_t firstpass_mb_stats; /**< first pass mb packet */
200 struct vpx_psnr_pkt {
201 unsigned int samples[4]; /**< Number of samples, total/y/u/v */
202 uint64_t sse[4]; /**< sum squared error, total/y/u/v */
203 double psnr[4]; /**< PSNR, total/y/u/v */
204 } psnr; /**< data for PSNR packet */
205 vpx_fixed_buf_t raw; /**< data for arbitrary packets */
206 // Spatial SVC is still experimental and may be removed before the next
207 // ABI bump.
208 #if VPX_ENCODER_ABI_VERSION > (4 + VPX_CODEC_ABI_VERSION)
209 size_t layer_sizes[VPX_SS_MAX_LAYERS];
210 struct vpx_psnr_pkt layer_psnr[VPX_SS_MAX_LAYERS];
211 #endif
213 /* This packet size is fixed to allow codecs to extend this
214 * interface without having to manage storage for raw packets,
215 * i.e., if it's smaller than 128 bytes, you can store in the
216 * packet list directly.
218 char pad[128 - sizeof(enum vpx_codec_cx_pkt_kind)]; /**< fixed sz */
219 } data; /**< packet data */
220 } vpx_codec_cx_pkt_t; /**< alias for struct vpx_codec_cx_pkt */
223 /*!\brief Encoder return output buffer callback
225 * This callback function, when registered, returns with packets when each
226 * spatial layer is encoded.
228 // putting the definitions here for now. (agrange: find if there
229 // is a better place for this)
230 typedef void (* vpx_codec_enc_output_cx_pkt_cb_fn_t)(vpx_codec_cx_pkt_t *pkt,
231 void *user_data);
233 /*!\brief Callback function pointer / user data pair storage */
234 typedef struct vpx_codec_enc_output_cx_cb_pair {
235 vpx_codec_enc_output_cx_pkt_cb_fn_t output_cx_pkt; /**< Callback function */
236 void *user_priv; /**< Pointer to private data */
237 } vpx_codec_priv_output_cx_pkt_cb_pair_t;
239 /*!\brief Rational Number
241 * This structure holds a fractional value.
243 typedef struct vpx_rational {
244 int num; /**< fraction numerator */
245 int den; /**< fraction denominator */
246 } vpx_rational_t; /**< alias for struct vpx_rational */
249 /*!\brief Multi-pass Encoding Pass */
250 enum vpx_enc_pass {
251 VPX_RC_ONE_PASS, /**< Single pass mode */
252 VPX_RC_FIRST_PASS, /**< First pass of multi-pass mode */
253 VPX_RC_LAST_PASS /**< Final pass of multi-pass mode */
257 /*!\brief Rate control mode */
258 enum vpx_rc_mode {
259 VPX_VBR, /**< Variable Bit Rate (VBR) mode */
260 VPX_CBR, /**< Constant Bit Rate (CBR) mode */
261 VPX_CQ, /**< Constrained Quality (CQ) mode */
262 VPX_Q, /**< Constant Quality (Q) mode */
266 /*!\brief Keyframe placement mode.
268 * This enumeration determines whether keyframes are placed automatically by
269 * the encoder or whether this behavior is disabled. Older releases of this
270 * SDK were implemented such that VPX_KF_FIXED meant keyframes were disabled.
271 * This name is confusing for this behavior, so the new symbols to be used
272 * are VPX_KF_AUTO and VPX_KF_DISABLED.
274 enum vpx_kf_mode {
275 VPX_KF_FIXED, /**< deprecated, implies VPX_KF_DISABLED */
276 VPX_KF_AUTO, /**< Encoder determines optimal placement automatically */
277 VPX_KF_DISABLED = 0 /**< Encoder does not place keyframes. */
281 /*!\brief Encoded Frame Flags
283 * This type indicates a bitfield to be passed to vpx_codec_encode(), defining
284 * per-frame boolean values. By convention, bits common to all codecs will be
285 * named VPX_EFLAG_*, and bits specific to an algorithm will be named
286 * /algo/_eflag_*. The lower order 16 bits are reserved for common use.
288 typedef long vpx_enc_frame_flags_t;
289 #define VPX_EFLAG_FORCE_KF (1<<0) /**< Force this frame to be a keyframe */
292 /*!\brief Encoder configuration structure
294 * This structure contains the encoder settings that have common representations
295 * across all codecs. This doesn't imply that all codecs support all features,
296 * however.
298 typedef struct vpx_codec_enc_cfg {
300 * generic settings (g)
303 /*!\brief Algorithm specific "usage" value
305 * Algorithms may define multiple values for usage, which may convey the
306 * intent of how the application intends to use the stream. If this value
307 * is non-zero, consult the documentation for the codec to determine its
308 * meaning.
310 unsigned int g_usage;
313 /*!\brief Maximum number of threads to use
315 * For multi-threaded implementations, use no more than this number of
316 * threads. The codec may use fewer threads than allowed. The value
317 * 0 is equivalent to the value 1.
319 unsigned int g_threads;
322 /*!\brief Bitstream profile to use
324 * Some codecs support a notion of multiple bitstream profiles. Typically
325 * this maps to a set of features that are turned on or off. Often the
326 * profile to use is determined by the features of the intended decoder.
327 * Consult the documentation for the codec to determine the valid values
328 * for this parameter, or set to zero for a sane default.
330 unsigned int g_profile; /**< profile of bitstream to use */
334 /*!\brief Width of the frame
336 * This value identifies the presentation resolution of the frame,
337 * in pixels. Note that the frames passed as input to the encoder must
338 * have this resolution. Frames will be presented by the decoder in this
339 * resolution, independent of any spatial resampling the encoder may do.
341 unsigned int g_w;
344 /*!\brief Height of the frame
346 * This value identifies the presentation resolution of the frame,
347 * in pixels. Note that the frames passed as input to the encoder must
348 * have this resolution. Frames will be presented by the decoder in this
349 * resolution, independent of any spatial resampling the encoder may do.
351 unsigned int g_h;
353 /*!\brief Bit-depth of the codec
355 * This value identifies the bit_depth of the codec,
356 * Only certain bit-depths are supported as identified in the
357 * vpx_bit_depth_t enum.
359 vpx_bit_depth_t g_bit_depth;
361 /*!\brief Bit-depth of the input frames
363 * This value identifies the bit_depth of the input frames in bits.
364 * Note that the frames passed as input to the encoder must have
365 * this bit-depth.
367 unsigned int g_input_bit_depth;
369 /*!\brief Stream timebase units
371 * Indicates the smallest interval of time, in seconds, used by the stream.
372 * For fixed frame rate material, or variable frame rate material where
373 * frames are timed at a multiple of a given clock (ex: video capture),
374 * the \ref RECOMMENDED method is to set the timebase to the reciprocal
375 * of the frame rate (ex: 1001/30000 for 29.970 Hz NTSC). This allows the
376 * pts to correspond to the frame number, which can be handy. For
377 * re-encoding video from containers with absolute time timestamps, the
378 * \ref RECOMMENDED method is to set the timebase to that of the parent
379 * container or multimedia framework (ex: 1/1000 for ms, as in FLV).
381 struct vpx_rational g_timebase;
384 /*!\brief Enable error resilient modes.
386 * The error resilient bitfield indicates to the encoder which features
387 * it should enable to take measures for streaming over lossy or noisy
388 * links.
390 vpx_codec_er_flags_t g_error_resilient;
393 /*!\brief Multi-pass Encoding Mode
395 * This value should be set to the current phase for multi-pass encoding.
396 * For single pass, set to #VPX_RC_ONE_PASS.
398 enum vpx_enc_pass g_pass;
401 /*!\brief Allow lagged encoding
403 * If set, this value allows the encoder to consume a number of input
404 * frames before producing output frames. This allows the encoder to
405 * base decisions for the current frame on future frames. This does
406 * increase the latency of the encoding pipeline, so it is not appropriate
407 * in all situations (ex: realtime encoding).
409 * Note that this is a maximum value -- the encoder may produce frames
410 * sooner than the given limit. Set this value to 0 to disable this
411 * feature.
413 unsigned int g_lag_in_frames;
417 * rate control settings (rc)
420 /*!\brief Temporal resampling configuration, if supported by the codec.
422 * Temporal resampling allows the codec to "drop" frames as a strategy to
423 * meet its target data rate. This can cause temporal discontinuities in
424 * the encoded video, which may appear as stuttering during playback. This
425 * trade-off is often acceptable, but for many applications is not. It can
426 * be disabled in these cases.
428 * Note that not all codecs support this feature. All vpx VPx codecs do.
429 * For other codecs, consult the documentation for that algorithm.
431 * This threshold is described as a percentage of the target data buffer.
432 * When the data buffer falls below this percentage of fullness, a
433 * dropped frame is indicated. Set the threshold to zero (0) to disable
434 * this feature.
436 unsigned int rc_dropframe_thresh;
439 /*!\brief Enable/disable spatial resampling, if supported by the codec.
441 * Spatial resampling allows the codec to compress a lower resolution
442 * version of the frame, which is then upscaled by the encoder to the
443 * correct presentation resolution. This increases visual quality at
444 * low data rates, at the expense of CPU time on the encoder/decoder.
446 unsigned int rc_resize_allowed;
448 /*!\brief Internal coded frame width.
450 * If spatial resampling is enabled this specifies the width of the
451 * encoded frame.
453 unsigned int rc_scaled_width;
455 /*!\brief Internal coded frame height.
457 * If spatial resampling is enabled this specifies the height of the
458 * encoded frame.
460 unsigned int rc_scaled_height;
462 /*!\brief Spatial resampling up watermark.
464 * This threshold is described as a percentage of the target data buffer.
465 * When the data buffer rises above this percentage of fullness, the
466 * encoder will step up to a higher resolution version of the frame.
468 unsigned int rc_resize_up_thresh;
471 /*!\brief Spatial resampling down watermark.
473 * This threshold is described as a percentage of the target data buffer.
474 * When the data buffer falls below this percentage of fullness, the
475 * encoder will step down to a lower resolution version of the frame.
477 unsigned int rc_resize_down_thresh;
480 /*!\brief Rate control algorithm to use.
482 * Indicates whether the end usage of this stream is to be streamed over
483 * a bandwidth constrained link, indicating that Constant Bit Rate (CBR)
484 * mode should be used, or whether it will be played back on a high
485 * bandwidth link, as from a local disk, where higher variations in
486 * bitrate are acceptable.
488 enum vpx_rc_mode rc_end_usage;
491 /*!\brief Two-pass stats buffer.
493 * A buffer containing all of the stats packets produced in the first
494 * pass, concatenated.
496 vpx_fixed_buf_t rc_twopass_stats_in;
498 /*!\brief first pass mb stats buffer.
500 * A buffer containing all of the first pass mb stats packets produced
501 * in the first pass, concatenated.
503 vpx_fixed_buf_t rc_firstpass_mb_stats_in;
505 /*!\brief Target data rate
507 * Target bandwidth to use for this stream, in kilobits per second.
509 unsigned int rc_target_bitrate;
513 * quantizer settings
517 /*!\brief Minimum (Best Quality) Quantizer
519 * The quantizer is the most direct control over the quality of the
520 * encoded image. The range of valid values for the quantizer is codec
521 * specific. Consult the documentation for the codec to determine the
522 * values to use. To determine the range programmatically, call
523 * vpx_codec_enc_config_default() with a usage value of 0.
525 unsigned int rc_min_quantizer;
528 /*!\brief Maximum (Worst Quality) Quantizer
530 * The quantizer is the most direct control over the quality of the
531 * encoded image. The range of valid values for the quantizer is codec
532 * specific. Consult the documentation for the codec to determine the
533 * values to use. To determine the range programmatically, call
534 * vpx_codec_enc_config_default() with a usage value of 0.
536 unsigned int rc_max_quantizer;
540 * bitrate tolerance
544 /*!\brief Rate control adaptation undershoot control
546 * This value, expressed as a percentage of the target bitrate,
547 * controls the maximum allowed adaptation speed of the codec.
548 * This factor controls the maximum amount of bits that can
549 * be subtracted from the target bitrate in order to compensate
550 * for prior overshoot.
552 * Valid values in the range 0-1000.
554 unsigned int rc_undershoot_pct;
557 /*!\brief Rate control adaptation overshoot control
559 * This value, expressed as a percentage of the target bitrate,
560 * controls the maximum allowed adaptation speed of the codec.
561 * This factor controls the maximum amount of bits that can
562 * be added to the target bitrate in order to compensate for
563 * prior undershoot.
565 * Valid values in the range 0-1000.
567 unsigned int rc_overshoot_pct;
571 * decoder buffer model parameters
575 /*!\brief Decoder Buffer Size
577 * This value indicates the amount of data that may be buffered by the
578 * decoding application. Note that this value is expressed in units of
579 * time (milliseconds). For example, a value of 5000 indicates that the
580 * client will buffer (at least) 5000ms worth of encoded data. Use the
581 * target bitrate (#rc_target_bitrate) to convert to bits/bytes, if
582 * necessary.
584 unsigned int rc_buf_sz;
587 /*!\brief Decoder Buffer Initial Size
589 * This value indicates the amount of data that will be buffered by the
590 * decoding application prior to beginning playback. This value is
591 * expressed in units of time (milliseconds). Use the target bitrate
592 * (#rc_target_bitrate) to convert to bits/bytes, if necessary.
594 unsigned int rc_buf_initial_sz;
597 /*!\brief Decoder Buffer Optimal Size
599 * This value indicates the amount of data that the encoder should try
600 * to maintain in the decoder's buffer. This value is expressed in units
601 * of time (milliseconds). Use the target bitrate (#rc_target_bitrate)
602 * to convert to bits/bytes, if necessary.
604 unsigned int rc_buf_optimal_sz;
608 * 2 pass rate control parameters
612 /*!\brief Two-pass mode CBR/VBR bias
614 * Bias, expressed on a scale of 0 to 100, for determining target size
615 * for the current frame. The value 0 indicates the optimal CBR mode
616 * value should be used. The value 100 indicates the optimal VBR mode
617 * value should be used. Values in between indicate which way the
618 * encoder should "lean."
620 unsigned int rc_2pass_vbr_bias_pct; /**< RC mode bias between CBR and VBR(0-100: 0->CBR, 100->VBR) */
623 /*!\brief Two-pass mode per-GOP minimum bitrate
625 * This value, expressed as a percentage of the target bitrate, indicates
626 * the minimum bitrate to be used for a single GOP (aka "section")
628 unsigned int rc_2pass_vbr_minsection_pct;
631 /*!\brief Two-pass mode per-GOP maximum bitrate
633 * This value, expressed as a percentage of the target bitrate, indicates
634 * the maximum bitrate to be used for a single GOP (aka "section")
636 unsigned int rc_2pass_vbr_maxsection_pct;
640 * keyframing settings (kf)
643 /*!\brief Keyframe placement mode
645 * This value indicates whether the encoder should place keyframes at a
646 * fixed interval, or determine the optimal placement automatically
647 * (as governed by the #kf_min_dist and #kf_max_dist parameters)
649 enum vpx_kf_mode kf_mode;
652 /*!\brief Keyframe minimum interval
654 * This value, expressed as a number of frames, prevents the encoder from
655 * placing a keyframe nearer than kf_min_dist to the previous keyframe. At
656 * least kf_min_dist frames non-keyframes will be coded before the next
657 * keyframe. Set kf_min_dist equal to kf_max_dist for a fixed interval.
659 unsigned int kf_min_dist;
662 /*!\brief Keyframe maximum interval
664 * This value, expressed as a number of frames, forces the encoder to code
665 * a keyframe if one has not been coded in the last kf_max_dist frames.
666 * A value of 0 implies all frames will be keyframes. Set kf_min_dist
667 * equal to kf_max_dist for a fixed interval.
669 unsigned int kf_max_dist;
672 * Spatial scalability settings (ss)
675 /*!\brief Number of spatial coding layers.
677 * This value specifies the number of spatial coding layers to be used.
679 unsigned int ss_number_layers;
681 /*!\brief Enable auto alt reference flags for each spatial layer.
683 * These values specify if auto alt reference frame is enabled for each
684 * spatial layer.
686 int ss_enable_auto_alt_ref[VPX_SS_MAX_LAYERS];
688 /*!\brief Target bitrate for each spatial layer.
690 * These values specify the target coding bitrate to be used for each
691 * spatial layer.
693 unsigned int ss_target_bitrate[VPX_SS_MAX_LAYERS];
695 /*!\brief Number of temporal coding layers.
697 * This value specifies the number of temporal layers to be used.
699 unsigned int ts_number_layers;
701 /*!\brief Target bitrate for each temporal layer.
703 * These values specify the target coding bitrate to be used for each
704 * temporal layer.
706 unsigned int ts_target_bitrate[VPX_TS_MAX_LAYERS];
708 /*!\brief Frame rate decimation factor for each temporal layer.
710 * These values specify the frame rate decimation factors to apply
711 * to each temporal layer.
713 unsigned int ts_rate_decimator[VPX_TS_MAX_LAYERS];
715 /*!\brief Length of the sequence defining frame temporal layer membership.
717 * This value specifies the length of the sequence that defines the
718 * membership of frames to temporal layers. For example, if the
719 * ts_periodicity = 8, then the frames are assigned to coding layers with a
720 * repeated sequence of length 8.
722 unsigned int ts_periodicity;
724 /*!\brief Template defining the membership of frames to temporal layers.
726 * This array defines the membership of frames to temporal coding layers.
727 * For a 2-layer encoding that assigns even numbered frames to one temporal
728 * layer (0) and odd numbered frames to a second temporal layer (1) with
729 * ts_periodicity=8, then ts_layer_id = (0,1,0,1,0,1,0,1).
731 unsigned int ts_layer_id[VPX_TS_MAX_PERIODICITY];
732 } vpx_codec_enc_cfg_t; /**< alias for struct vpx_codec_enc_cfg */
734 /*!\brief vp9 svc extra configure parameters
736 * This defines max/min quantizers and scale factors for each layer
739 typedef struct vpx_svc_parameters {
740 int max_quantizers[VPX_SS_MAX_LAYERS]; /**< Max Q for each layer */
741 int min_quantizers[VPX_SS_MAX_LAYERS]; /**< Min Q for each layer */
742 int scaling_factor_num[VPX_SS_MAX_LAYERS]; /**< Scaling factor-numerator*/
743 int scaling_factor_den[VPX_SS_MAX_LAYERS]; /**< Scaling factor-denominator*/
744 } vpx_svc_extra_cfg_t;
747 /*!\brief Initialize an encoder instance
749 * Initializes a encoder context using the given interface. Applications
750 * should call the vpx_codec_enc_init convenience macro instead of this
751 * function directly, to ensure that the ABI version number parameter
752 * is properly initialized.
754 * If the library was configured with --disable-multithread, this call
755 * is not thread safe and should be guarded with a lock if being used
756 * in a multithreaded context.
758 * \param[in] ctx Pointer to this instance's context.
759 * \param[in] iface Pointer to the algorithm interface to use.
760 * \param[in] cfg Configuration to use, if known. May be NULL.
761 * \param[in] flags Bitfield of VPX_CODEC_USE_* flags
762 * \param[in] ver ABI version number. Must be set to
763 * VPX_ENCODER_ABI_VERSION
764 * \retval #VPX_CODEC_OK
765 * The decoder algorithm initialized.
766 * \retval #VPX_CODEC_MEM_ERROR
767 * Memory allocation failed.
769 vpx_codec_err_t vpx_codec_enc_init_ver(vpx_codec_ctx_t *ctx,
770 vpx_codec_iface_t *iface,
771 const vpx_codec_enc_cfg_t *cfg,
772 vpx_codec_flags_t flags,
773 int ver);
776 /*!\brief Convenience macro for vpx_codec_enc_init_ver()
778 * Ensures the ABI version parameter is properly set.
780 #define vpx_codec_enc_init(ctx, iface, cfg, flags) \
781 vpx_codec_enc_init_ver(ctx, iface, cfg, flags, VPX_ENCODER_ABI_VERSION)
784 /*!\brief Initialize multi-encoder instance
786 * Initializes multi-encoder context using the given interface.
787 * Applications should call the vpx_codec_enc_init_multi convenience macro
788 * instead of this function directly, to ensure that the ABI version number
789 * parameter is properly initialized.
791 * \param[in] ctx Pointer to this instance's context.
792 * \param[in] iface Pointer to the algorithm interface to use.
793 * \param[in] cfg Configuration to use, if known. May be NULL.
794 * \param[in] num_enc Total number of encoders.
795 * \param[in] flags Bitfield of VPX_CODEC_USE_* flags
796 * \param[in] dsf Pointer to down-sampling factors.
797 * \param[in] ver ABI version number. Must be set to
798 * VPX_ENCODER_ABI_VERSION
799 * \retval #VPX_CODEC_OK
800 * The decoder algorithm initialized.
801 * \retval #VPX_CODEC_MEM_ERROR
802 * Memory allocation failed.
804 vpx_codec_err_t vpx_codec_enc_init_multi_ver(vpx_codec_ctx_t *ctx,
805 vpx_codec_iface_t *iface,
806 vpx_codec_enc_cfg_t *cfg,
807 int num_enc,
808 vpx_codec_flags_t flags,
809 vpx_rational_t *dsf,
810 int ver);
813 /*!\brief Convenience macro for vpx_codec_enc_init_multi_ver()
815 * Ensures the ABI version parameter is properly set.
817 #define vpx_codec_enc_init_multi(ctx, iface, cfg, num_enc, flags, dsf) \
818 vpx_codec_enc_init_multi_ver(ctx, iface, cfg, num_enc, flags, dsf, \
819 VPX_ENCODER_ABI_VERSION)
822 /*!\brief Get a default configuration
824 * Initializes a encoder configuration structure with default values. Supports
825 * the notion of "usages" so that an algorithm may offer different default
826 * settings depending on the user's intended goal. This function \ref SHOULD
827 * be called by all applications to initialize the configuration structure
828 * before specializing the configuration with application specific values.
830 * \param[in] iface Pointer to the algorithm interface to use.
831 * \param[out] cfg Configuration buffer to populate.
832 * \param[in] reserved Must set to 0 for VP8 and VP9.
834 * \retval #VPX_CODEC_OK
835 * The configuration was populated.
836 * \retval #VPX_CODEC_INCAPABLE
837 * Interface is not an encoder interface.
838 * \retval #VPX_CODEC_INVALID_PARAM
839 * A parameter was NULL, or the usage value was not recognized.
841 vpx_codec_err_t vpx_codec_enc_config_default(vpx_codec_iface_t *iface,
842 vpx_codec_enc_cfg_t *cfg,
843 unsigned int reserved);
846 /*!\brief Set or change configuration
848 * Reconfigures an encoder instance according to the given configuration.
850 * \param[in] ctx Pointer to this instance's context
851 * \param[in] cfg Configuration buffer to use
853 * \retval #VPX_CODEC_OK
854 * The configuration was populated.
855 * \retval #VPX_CODEC_INCAPABLE
856 * Interface is not an encoder interface.
857 * \retval #VPX_CODEC_INVALID_PARAM
858 * A parameter was NULL, or the usage value was not recognized.
860 vpx_codec_err_t vpx_codec_enc_config_set(vpx_codec_ctx_t *ctx,
861 const vpx_codec_enc_cfg_t *cfg);
864 /*!\brief Get global stream headers
866 * Retrieves a stream level global header packet, if supported by the codec.
868 * \param[in] ctx Pointer to this instance's context
870 * \retval NULL
871 * Encoder does not support global header
872 * \retval Non-NULL
873 * Pointer to buffer containing global header packet
875 vpx_fixed_buf_t *vpx_codec_get_global_headers(vpx_codec_ctx_t *ctx);
878 #define VPX_DL_REALTIME (1) /**< deadline parameter analogous to
879 * VPx REALTIME mode. */
880 #define VPX_DL_GOOD_QUALITY (1000000) /**< deadline parameter analogous to
881 * VPx GOOD QUALITY mode. */
882 #define VPX_DL_BEST_QUALITY (0) /**< deadline parameter analogous to
883 * VPx BEST QUALITY mode. */
884 /*!\brief Encode a frame
886 * Encodes a video frame at the given "presentation time." The presentation
887 * time stamp (PTS) \ref MUST be strictly increasing.
889 * The encoder supports the notion of a soft real-time deadline. Given a
890 * non-zero value to the deadline parameter, the encoder will make a "best
891 * effort" guarantee to return before the given time slice expires. It is
892 * implicit that limiting the available time to encode will degrade the
893 * output quality. The encoder can be given an unlimited time to produce the
894 * best possible frame by specifying a deadline of '0'. This deadline
895 * supercedes the VPx notion of "best quality, good quality, realtime".
896 * Applications that wish to map these former settings to the new deadline
897 * based system can use the symbols #VPX_DL_REALTIME, #VPX_DL_GOOD_QUALITY,
898 * and #VPX_DL_BEST_QUALITY.
900 * When the last frame has been passed to the encoder, this function should
901 * continue to be called, with the img parameter set to NULL. This will
902 * signal the end-of-stream condition to the encoder and allow it to encode
903 * any held buffers. Encoding is complete when vpx_codec_encode() is called
904 * and vpx_codec_get_cx_data() returns no data.
906 * \param[in] ctx Pointer to this instance's context
907 * \param[in] img Image data to encode, NULL to flush.
908 * \param[in] pts Presentation time stamp, in timebase units.
909 * \param[in] duration Duration to show frame, in timebase units.
910 * \param[in] flags Flags to use for encoding this frame.
911 * \param[in] deadline Time to spend encoding, in microseconds. (0=infinite)
913 * \retval #VPX_CODEC_OK
914 * The configuration was populated.
915 * \retval #VPX_CODEC_INCAPABLE
916 * Interface is not an encoder interface.
917 * \retval #VPX_CODEC_INVALID_PARAM
918 * A parameter was NULL, the image format is unsupported, etc.
920 vpx_codec_err_t vpx_codec_encode(vpx_codec_ctx_t *ctx,
921 const vpx_image_t *img,
922 vpx_codec_pts_t pts,
923 unsigned long duration,
924 vpx_enc_frame_flags_t flags,
925 unsigned long deadline);
927 /*!\brief Set compressed data output buffer
929 * Sets the buffer that the codec should output the compressed data
930 * into. This call effectively sets the buffer pointer returned in the
931 * next VPX_CODEC_CX_FRAME_PKT packet. Subsequent packets will be
932 * appended into this buffer. The buffer is preserved across frames,
933 * so applications must periodically call this function after flushing
934 * the accumulated compressed data to disk or to the network to reset
935 * the pointer to the buffer's head.
937 * `pad_before` bytes will be skipped before writing the compressed
938 * data, and `pad_after` bytes will be appended to the packet. The size
939 * of the packet will be the sum of the size of the actual compressed
940 * data, pad_before, and pad_after. The padding bytes will be preserved
941 * (not overwritten).
943 * Note that calling this function does not guarantee that the returned
944 * compressed data will be placed into the specified buffer. In the
945 * event that the encoded data will not fit into the buffer provided,
946 * the returned packet \ref MAY point to an internal buffer, as it would
947 * if this call were never used. In this event, the output packet will
948 * NOT have any padding, and the application must free space and copy it
949 * to the proper place. This is of particular note in configurations
950 * that may output multiple packets for a single encoded frame (e.g., lagged
951 * encoding) or if the application does not reset the buffer periodically.
953 * Applications may restore the default behavior of the codec providing
954 * the compressed data buffer by calling this function with a NULL
955 * buffer.
957 * Applications \ref MUSTNOT call this function during iteration of
958 * vpx_codec_get_cx_data().
960 * \param[in] ctx Pointer to this instance's context
961 * \param[in] buf Buffer to store compressed data into
962 * \param[in] pad_before Bytes to skip before writing compressed data
963 * \param[in] pad_after Bytes to skip after writing compressed data
965 * \retval #VPX_CODEC_OK
966 * The buffer was set successfully.
967 * \retval #VPX_CODEC_INVALID_PARAM
968 * A parameter was NULL, the image format is unsupported, etc.
970 vpx_codec_err_t vpx_codec_set_cx_data_buf(vpx_codec_ctx_t *ctx,
971 const vpx_fixed_buf_t *buf,
972 unsigned int pad_before,
973 unsigned int pad_after);
976 /*!\brief Encoded data iterator
978 * Iterates over a list of data packets to be passed from the encoder to the
979 * application. The different kinds of packets available are enumerated in
980 * #vpx_codec_cx_pkt_kind.
982 * #VPX_CODEC_CX_FRAME_PKT packets should be passed to the application's
983 * muxer. Multiple compressed frames may be in the list.
984 * #VPX_CODEC_STATS_PKT packets should be appended to a global buffer.
986 * The application \ref MUST silently ignore any packet kinds that it does
987 * not recognize or support.
989 * The data buffers returned from this function are only guaranteed to be
990 * valid until the application makes another call to any vpx_codec_* function.
992 * \param[in] ctx Pointer to this instance's context
993 * \param[in,out] iter Iterator storage, initialized to NULL
995 * \return Returns a pointer to an output data packet (compressed frame data,
996 * two-pass statistics, etc.) or NULL to signal end-of-list.
999 const vpx_codec_cx_pkt_t *vpx_codec_get_cx_data(vpx_codec_ctx_t *ctx,
1000 vpx_codec_iter_t *iter);
1003 /*!\brief Get Preview Frame
1005 * Returns an image that can be used as a preview. Shows the image as it would
1006 * exist at the decompressor. The application \ref MUST NOT write into this
1007 * image buffer.
1009 * \param[in] ctx Pointer to this instance's context
1011 * \return Returns a pointer to a preview image, or NULL if no image is
1012 * available.
1015 const vpx_image_t *vpx_codec_get_preview_frame(vpx_codec_ctx_t *ctx);
1018 /*!@} - end defgroup encoder*/
1019 #ifdef __cplusplus
1021 #endif
1022 #endif // VPX_VPX_ENCODER_H_