Searched refs:quantizer (Results 1 – 15 of 15) sorted by relevance
3 76024eb753cdac6a5e5703aaea189d35c3c30ac7 *invalid-vp90-2-00-quantizer-00.webm.ivf.s5861_r01-05_b6-.…4 7448d8798a4380162d4b56f9b452e2f6f9e24e7a *invalid-vp90-2-00-quantizer-00.webm.ivf.s5861_r01-05_b6-.…5 83f50908c8dc0ef8760595447a2ff7727489542e *invalid-vp90-2-00-quantizer-11.webm.ivf.s52984_r01-05_b6-…6 456d1493e52d32a5c30edf44a27debc1fa6b253a *invalid-vp90-2-00-quantizer-11.webm.ivf.s52984_r01-05_b6-…7 c123d1f9f02fb4143abb5e271916e3a3080de8f6 *invalid-vp90-2-00-quantizer-11.webm.ivf.s52984_r01-05_b6-…8 456d1493e52d32a5c30edf44a27debc1fa6b253a *invalid-vp90-2-00-quantizer-11.webm.ivf.s52984_r01-05_b6-…158 ce881e567fe1d0fbcb2d3e9e6281a1a8d74d82e0 *vp90-2-00-quantizer-00.webm159 ac5eda33407d0521c7afca43a63fd305c0cd9d13 *vp90-2-00-quantizer-00.webm.md5160 2ca0463f2cfb93d25d7dded174db70b7cb87cb48 *vp90-2-00-quantizer-01.webm161 10d98884fc6d9a5f47a2057922b8e25dd48d7786 *vp90-2-00-quantizer-01.webm.md5[all …]
153 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-00.webm154 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-00.webm.md5155 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-01.webm156 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-01.webm.md5157 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-02.webm158 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-02.webm.md5159 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-03.webm160 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-03.webm.md5161 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-04.webm162 LIBVPX_TEST_DATA-$(CONFIG_VP9_DECODER) += vp90-2-00-quantizer-04.webm.md5[all …]
11 The fake quantization operator achieves this by modeling the quantizer as a pass52 For the backward pass, we model the quantizer as a piecewise linear block, with61 Therefore, the backward pass through the quantizer reduces to passing through62 the gradients as long as the inputs to the quantizer are in the linear region.65 Note that the quantizer is fully specified by the min and max values of the
339 int vp9_quantizer_to_qindex(int quantizer) { in vp9_quantizer_to_qindex() argument340 return quantizer_to_qindex[quantizer]; in vp9_quantizer_to_qindex()344 int quantizer; in vp9_qindex_to_quantizer() local346 for (quantizer = 0; quantizer < 64; ++quantizer) in vp9_qindex_to_quantizer()347 if (quantizer_to_qindex[quantizer] >= qindex) return quantizer; in vp9_qindex_to_quantizer()
54 int vp9_quantizer_to_qindex(int quantizer);
248 int quantizer = qstep_vec[i]; in model_rd_for_sb() local250 if (quantizer < 120) in model_rd_for_sb()251 rate = (square_error * (280 - quantizer)) >> (16 - VP9_PROB_COST_SHIFT); in model_rd_for_sb()254 dist = (square_error * quantizer) >> 8; in model_rd_for_sb()
50 context->desc.vc1.quantizer = vc1->pic_quantizer_fields.bits.quantizer; in vlVaHandlePictureParameterBufferVC1()
71 uint8_t quantizer; // 13 member183 vc->quantizer = d->quantizer; in nouveau_vp3_fill_picparm_vc1_bsp()
107 uint8_t quantizer; // 35 member449 vc->quantizer = d->quantizer; in nouveau_vp3_fill_picparm_vc1_vp()
374 vpxenc: add quantizer histogram display472 SSSE3 version of fast quantizer493 Tuning for the more exact quantizer.496 Limit key frame quantizer for forced key frames.499 make rdmult adaptive for intra in quantizer RDO500 cap the best quantizer for 2nd order DC514 quantizer: fix assertion in fast quantizer path625 Improved control of active min quantizer for two pass.647 Normalize quantizer's zero bin and rounding factors649 Make the quantizer exact.
121 FIXP_QTD quantizer = FDKaacEnc_quantTableQ[(-gain) & 3]; in FDKaacEnc_quantizeLines() local131 FIXP_DBL accu = fMultDiv2(mdctSpectrum[line], quantizer); in FDKaacEnc_quantizeLines()
252 uint8_t quantizer; member
333 picture->quantizer = picture_info->quantizer; in vlVdpDecoderRenderVC1()
773 result.pps_info_flags |= pic->quantizer << 9; in get_vc1_msg()
1410 The two-pass color quantizer, jquant2.c, is specialized to handle RGB data2214 IMPORTANT: because the different quantizer methods have very different2233 quantizer to be re-run to generate a new colormap.