0DOF split rendering support for non-diegetic stereo out (a4cacd75) · Commits · IVAS Codec Public Collaboration / IVAS Codec

apps/decoder.c

+11 −9

Original line number	Diff line number	Diff line
		@@ -417,7 +417,8 @@ int main(
		/* sanity check */
		if ( arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_ROOM_IR && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_ROOM_REVERB
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		&& arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM
		&& arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM &&
		arg.Opt_non_diegetic_pan == 0
		#endif
		)
		{
		@@ -608,7 +609,8 @@ int main(
		/* sanity check */
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_ROOM_IR && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_ROOM_REVERB &&
		arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )
		arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED && arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM &&
		arg.Opt_non_diegetic_pan == 0 )
		{
		fprintf( stderr, "\nExternal Renderer Config is supported only when binaural output configurations is used as output OR when Split rendering mode is enabled. Exiting. \n" );
		goto cleanup;
		@@ -1800,7 +1802,7 @@ static ivas_error initOnFirstGoodFrame(
		}

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )
		if ( IVAS_DEC_is_split_rendering_enabled( hIvasDec ) )
		{
		pFullDelayNumSamples[0] = 0;
		}
		@@ -1827,7 +1829,7 @@ static ivas_error initOnFirstGoodFrame(
		}

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )
		if ( IVAS_DEC_is_split_rendering_enabled( hIvasDec ) )
		{
		/* Open split rendering metadata writer */
		int16_t delayNumSamples_temp[3];
		@@ -1844,7 +1846,7 @@ static ivas_error initOnFirstGoodFrame(

		IVAS_DEC_GetSplitRendBitstreamHeader( hIvasDec, &splitRendCodec, &poseCorrection, &splitRendCodecFrameSizeMs );

		if ( arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED )
		if ( IVAS_DEC_is_split_rendering_coded_out( hIvasDec ) )
		{
		if ( ( error = split_rend_writer_open( splitRendWriter, arg.outputWavFilename, delayNumSamples_temp[0], delayTimeScale_temp, splitRendCodec, poseCorrection, splitRendCodecFrameSizeMs ) ) != IVAS_ERR_OK )
		{
		@@ -1862,7 +1864,7 @@ static ivas_error initOnFirstGoodFrame(
		}
		}

		if ( arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED )
		if ( IVAS_DEC_is_split_rendering_coded_out( hIvasDec ) == 0 )
		{
		#endif
		/* Open audio writer and write all previously skipped bad frames now that frame size is known */
		@@ -2360,7 +2362,7 @@ static ivas_error decodeG192(
		}

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )
		if ( IVAS_DEC_is_split_rendering_enabled( hIvasDec ) )
		{
		if ( ( error = IVAS_DEC_GetSplitBinauralBitstream( hIvasDec, (void ) ( pcmBuf + nOutChannels nSamplesRendered ), splitRendBits, &nSamplesRendered_loop, &needNewFrame ) ) != IVAS_ERR_OK )
		{
		@@ -2442,7 +2444,7 @@ static ivas_error decodeG192(
		if ( decodedGoodFrame )
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| arg.outputConfig == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )
		if ( IVAS_DEC_is_split_rendering_enabled( hIvasDec ) )
		{
		if ( split_rend_write_bitstream_to_file( splitRendWriter, splitRendBits->bits_buf, &splitRendBits->bits_read, &splitRendBits->bits_written ) != IVAS_ERR_OK )
		{
		@@ -2451,7 +2453,7 @@ static ivas_error decodeG192(
		}
		}

		if ( arg.outputConfig != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED )
		if ( IVAS_DEC_is_split_rendering_coded_out( hIvasDec ) == 0 )
		{
		#endif
		if ( delayNumSamples < nOutSamples )

lib_dec/ivas_init_dec.c

+10 −4

Original line number	Diff line number	Diff line
		@@ -92,7 +92,7 @@ static ivas_error ivas_dec_reconfig_split_rend(
		cldfb_in_flag = 1;
		}

		isar_renderSplitGetMultiBinPoseData( &st_ivas->hRenderConfig->split_rend_config, &hSplitRendWrapper->multiBinPoseData, st_ivas->hHeadTrackData->sr_pose_pred_axis );
		isar_renderSplitGetMultiBinPoseData( &st_ivas->hRenderConfig->split_rend_config, &hSplitRendWrapper->multiBinPoseData, ( st_ivas->hHeadTrackData != NULL ) ? st_ivas->hHeadTrackData->sr_pose_pred_axis : DEFAULT_AXIS );

		isCldfbNeeded = 0;
		#ifndef SPLIT_REND_WITH_HEAD_ROT
		@@ -128,6 +128,10 @@ static ivas_error ivas_dec_reconfig_split_rend(
		#endif
		}
		}
		else if ( st_ivas->hDecoderConfig->Opt_non_diegetic_pan )
		{
		isCldfbNeeded = 1;
		}

		if ( isCldfbNeeded == 1 && hSplitRendWrapper->hCldfbHandles == NULL )
		{
		@@ -239,7 +243,7 @@ static ivas_error ivas_dec_init_split_rend(
		cldfb_in_flag = 1;
		}

		isar_renderSplitGetMultiBinPoseData( &st_ivas->hRenderConfig->split_rend_config, &st_ivas->hSplitBinRend.splitrend.multiBinPoseData, st_ivas->hHeadTrackData->sr_pose_pred_axis );
		isar_renderSplitGetMultiBinPoseData( &st_ivas->hRenderConfig->split_rend_config, &st_ivas->hSplitBinRend.splitrend.multiBinPoseData, ( st_ivas->hHeadTrackData != NULL ) ? st_ivas->hHeadTrackData->sr_pose_pred_axis : DEFAULT_AXIS );

		if ( cldfb_in_flag == 1 && ( st_ivas->hSplitBinRend.splitrend.multiBinPoseData.poseCorrectionMode == ISAR_SPLIT_REND_POSE_CORRECTION_MODE_NONE ) )
		{
		@@ -1095,7 +1099,8 @@ ivas_error ivas_init_decoder_front(

		if ( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_ROOM_IR \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_ROOM_REVERB
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		\|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM
		\|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM \|\|
		( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_STEREO && st_ivas->hDecoderConfig->Opt_non_diegetic_pan )
		#endif
		)
		{
		@@ -1257,7 +1262,8 @@ ivas_error ivas_init_decoder(
		* Initialize binuaral split rendering
		-----------------------------------------------------------------/

		if ( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )
		if ( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM \|\|
		( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_STEREO && st_ivas->hDecoderConfig->Opt_non_diegetic_pan && st_ivas->hRenderConfig->split_rend_config.dof == 0 ) )
		{
		if ( ( error = ivas_dec_init_split_rend( st_ivas ) ) != IVAS_ERR_OK )
		{

lib_dec/lib_dec.c

+60 −2

Original line number	Diff line number	Diff line
		@@ -1073,6 +1073,7 @@ ivas_error IVAS_DEC_GetSplitBinauralBitstream(
		int16_t numPoses;
		int16_t slots_rendered, slots_rendered_new;
		int16_t ro_md_flag;
		IVAS_QUATERNION Quaternion;

		error = IVAS_ERR_OK;
		st_ivas = hIvasDec->st_ivas;
		@@ -1097,7 +1098,8 @@ ivas_error IVAS_DEC_GetSplitBinauralBitstream(
		numSamplesPerChannelToDecode = (int16_t) ( output_Fs / FRAMES_PER_SEC / MAX_PARAM_SPATIAL_SUBFRAMES );
		numSamplesPerChannelToDecode *= (int16_t) st_ivas->hDecoderConfig->render_framesize;
		}
		if ( output_config != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED && output_config != IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM )

		if ( IVAS_DEC_is_split_rendering_enabled( hIvasDec ) == 0 )
		{
		return IVAS_ERR_WRONG_PARAMS;
		}
		@@ -1162,8 +1164,19 @@ ivas_error IVAS_DEC_GetSplitBinauralBitstream(
		ro_md_flag = 0;
		}

		if ( st_ivas->hHeadTrackData != NULL )
		{
		Quaternion = st_ivas->hHeadTrackData->Quaternions[0];
		}
		else
		{
		Quaternion.w = -3.0f;
		Quaternion.x = 0.0f;
		Quaternion.y = 0.0f;
		Quaternion.z = 0.0f;
		}
		if ( ( error = isar_renderMultiBinToSplitBinaural( &hSplitBinRend->splitrend,
		st_ivas->hHeadTrackData->Quaternions[0],
		Quaternion,
		st_ivas->hRenderConfig->split_rend_config.splitRendBitRate,
		st_ivas->hRenderConfig->split_rend_config.codec,
		st_ivas->hRenderConfig->split_rend_config.codec_frame_size_ms,
		@@ -3833,4 +3846,49 @@ PCM_RESOLUTION pcm_type_API_to_internal(
		return pcm_resolution;
		}

		int16_t IVAS_DEC_is_split_rendering_enabled(
		IVAS_DEC_HANDLE hIvasDec /* i/o: IVAS decoder handle */
		)
		{
		Decoder_Struct *st_ivas;
		int16_t isSplitRend;

		if ( hIvasDec == NULL \|\| hIvasDec->st_ivas == NULL )
		{
		return IVAS_ERR_UNEXPECTED_NULL_POINTER;
		}
		st_ivas = hIvasDec->st_ivas;
		isSplitRend = 0;

		if ( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\| st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_PCM \|\|
		( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_STEREO && st_ivas->hDecoderConfig->Opt_non_diegetic_pan && st_ivas->hRenderConfig->split_rend_config.dof == 0 ) )
		{
		isSplitRend = 1;
		}

		return isSplitRend;
		}

		int16_t IVAS_DEC_is_split_rendering_coded_out(
		IVAS_DEC_HANDLE hIvasDec /* i/o: IVAS decoder handle */
		)
		{
		Decoder_Struct *st_ivas;
		int16_t isSplitCoded;

		if ( hIvasDec == NULL \|\| hIvasDec->st_ivas == NULL )
		{
		return IVAS_ERR_UNEXPECTED_NULL_POINTER;
		}
		st_ivas = hIvasDec->st_ivas;
		isSplitCoded = 0;

		if ( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_BINAURAL_SPLIT_CODED \|\|
		( st_ivas->hDecoderConfig->output_config == IVAS_AUDIO_CONFIG_STEREO && st_ivas->hDecoderConfig->Opt_non_diegetic_pan && st_ivas->hRenderConfig->split_rend_config.dof == 0 ) )
		{
		isSplitCoded = 1;
		}

		return isSplitCoded;
		}
		#endif

lib_dec/lib_dec.h

+7 −0

Original line number	Diff line number	Diff line
		@@ -193,6 +193,13 @@ ivas_error IVAS_DEC_GetCldfbSamples(
		IVAS_AUDIO_CONFIG audio_config, / o : audio configuration */
		int16_t nOutSamples / o : number of samples per channel written to output buffer */
		);

		int16_t IVAS_DEC_is_split_rendering_enabled(
		IVAS_DEC_HANDLE hIvasDec /* i: IVAS decoder handle */
		);
		int16_t IVAS_DEC_is_split_rendering_coded_out(
		IVAS_DEC_HANDLE hIvasDec /* i/o: IVAS decoder handle */
		);
		#endif

		/! r: error code /

lib_isar/isar_splitRendererPre.c

+1 −1

Original line number	Diff line number	Diff line
		@@ -1661,7 +1661,7 @@ ivas_error isar_set_split_rend_setup(
		return IVAS_ERROR( IVAS_ERR_FAILED_ALLOC, "Cannot allocate memory for split rendering structure" );
		}

		isar_renderSplitGetMultiBinPoseData( hSplitBinConfig, &hSplitBinRend->splitrend.multiBinPoseData, hCombinedOrientationData->sr_pose_pred_axis );
		isar_renderSplitGetMultiBinPoseData( hSplitBinConfig, &hSplitBinRend->splitrend.multiBinPoseData, ( hCombinedOrientationData != NULL ) ? hCombinedOrientationData->sr_pose_pred_axis : DEFAULT_AXIS );

		if ( hCombinedOrientationData != NULL && hSplitBinRend->splitrend.multiBinPoseData.poseCorrectionMode == ISAR_SPLIT_REND_POSE_CORRECTION_MODE_CLDFB )
		{