arm/neon/denoising_neon.c

/*
 *  Copyright (c) 2012 The WebM project authors. All Rights Reserved.
 *
 *  Use of this source code is governed by a BSD-style license
 *  that can be found in the LICENSE file in the root of the source
 *  tree. An additional intellectual property rights grant can be found
 *  in the file PATENTS.  All contributing project authors may
 *  be found in the AUTHORS file in the root of the source tree.
 */

#include <arm_neon.h>

#include "vp8/encoder/denoising.h"
#include "vpx_mem/vpx_mem.h"
#include "./vp8_rtcd.h"

/*
 * The filter function was modified to reduce the computational complexity.
 *
 * Step 1:
 *  Instead of applying tap coefficients for each pixel, we calculated the
 *  pixel adjustments vs. pixel diff value ahead of time.
 *     adjustment = filtered_value - current_raw
 *                = (filter_coefficient * diff + 128) >> 8
 *  where
 *     filter_coefficient = (255 << 8) / (256 + ((abs_diff * 330) >> 3));
 *     filter_coefficient += filter_coefficient /
 *                           (3 + motion_magnitude_adjustment);
 *     filter_coefficient is clamped to 0 ~ 255.
 *
 * Step 2:
 *  The adjustment vs. diff curve becomes flat very quick when diff increases.
 *  This allowed us to use only several levels to approximate the curve without
 *  changing the filtering algorithm too much.
 *  The adjustments were further corrected by checking the motion magnitude.
 *  The levels used are:
 *      diff          level       adjustment w/o       adjustment w/
 *                               motion correction    motion correction
 *      [-255, -16]     3              -6                   -7
 *      [-15, -8]       2              -4                   -5
 *      [-7, -4]        1              -3                   -4
 *      [-3, 3]         0              diff                 diff
 *      [4, 7]          1               3                    4
 *      [8, 15]         2               4                    5
 *      [16, 255]       3               6                    7
 */

int vp8_denoiser_filter_neon(unsigned char *mc_running_avg_y,
                             int mc_running_avg_y_stride,
                             unsigned char *running_avg_y,
                             int running_avg_y_stride, unsigned char *sig,
                             int sig_stride, unsigned int motion_magnitude,
                             int increase_denoising) {
  /* If motion_magnitude is small, making the denoiser more aggressive by
   * increasing the adjustment for each level, level1 adjustment is
   * increased, the deltas stay the same.
   */
  int shift_inc =
      (increase_denoising && motion_magnitude <= MOTION_MAGNITUDE_THRESHOLD)
          ? 1
          : 0;
  const uint8x16_t v_level1_adjustment = vmovq_n_u8(
      (motion_magnitude <= MOTION_MAGNITUDE_THRESHOLD) ? 4 + shift_inc : 3);
  const uint8x16_t v_delta_level_1_and_2 = vdupq_n_u8(1);
  const uint8x16_t v_delta_level_2_and_3 = vdupq_n_u8(2);
  const uint8x16_t v_level1_threshold = vmovq_n_u8(4 + shift_inc);
  const uint8x16_t v_level2_threshold = vdupq_n_u8(8);
  const uint8x16_t v_level3_threshold = vdupq_n_u8(16);
  int64x2_t v_sum_diff_total = vdupq_n_s64(0);

  /* Go over lines. */
  int r;
  for (r = 0; r < 16; ++r) {
    /* Load inputs. */
    const uint8x16_t v_sig = vld1q_u8(sig);
    const uint8x16_t v_mc_running_avg_y = vld1q_u8(mc_running_avg_y);

    /* Calculate absolute difference and sign masks. */
    const uint8x16_t v_abs_diff = vabdq_u8(v_sig, v_mc_running_avg_y);
    const uint8x16_t v_diff_pos_mask = vcltq_u8(v_sig, v_mc_running_avg_y);
    const uint8x16_t v_diff_neg_mask = vcgtq_u8(v_sig, v_mc_running_avg_y);

    /* Figure out which level that put us in. */
    const uint8x16_t v_level1_mask = vcleq_u8(v_level1_threshold, v_abs_diff);
    const uint8x16_t v_level2_mask = vcleq_u8(v_level2_threshold, v_abs_diff);
    const uint8x16_t v_level3_mask = vcleq_u8(v_level3_threshold, v_abs_diff);

    /* Calculate absolute adjustments for level 1, 2 and 3. */
    const uint8x16_t v_level2_adjustment =
        vandq_u8(v_level2_mask, v_delta_level_1_and_2);
    const uint8x16_t v_level3_adjustment =
        vandq_u8(v_level3_mask, v_delta_level_2_and_3);
    const uint8x16_t v_level1and2_adjustment =
        vaddq_u8(v_level1_adjustment, v_level2_adjustment);
    const uint8x16_t v_level1and2and3_adjustment =
        vaddq_u8(v_level1and2_adjustment, v_level3_adjustment);

    /* Figure adjustment absolute value by selecting between the absolute
     * difference if in level0 or the value for level 1, 2 and 3.
     */
    const uint8x16_t v_abs_adjustment =
        vbslq_u8(v_level1_mask, v_level1and2and3_adjustment, v_abs_diff);

    /* Calculate positive and negative adjustments. Apply them to the signal
     * and accumulate them. Adjustments are less than eight and the maximum
     * sum of them (7 * 16) can fit in a signed char.
     */
    const uint8x16_t v_pos_adjustment =
        vandq_u8(v_diff_pos_mask, v_abs_adjustment);
    const uint8x16_t v_neg_adjustment =
        vandq_u8(v_diff_neg_mask, v_abs_adjustment);

    uint8x16_t v_running_avg_y = vqaddq_u8(v_sig, v_pos_adjustment);
    v_running_avg_y = vqsubq_u8(v_running_avg_y, v_neg_adjustment);

    /* Store results. */
    vst1q_u8(running_avg_y, v_running_avg_y);

    /* Sum all the accumulators to have the sum of all pixel differences
     * for this macroblock.
     */
    {
      const int8x16_t v_sum_diff =
          vqsubq_s8(vreinterpretq_s8_u8(v_pos_adjustment),
                    vreinterpretq_s8_u8(v_neg_adjustment));

      const int16x8_t fe_dc_ba_98_76_54_32_10 = vpaddlq_s8(v_sum_diff);

      const int32x4_t fedc_ba98_7654_3210 =
          vpaddlq_s16(fe_dc_ba_98_76_54_32_10);

      const int64x2_t fedcba98_76543210 = vpaddlq_s32(fedc_ba98_7654_3210);

      v_sum_diff_total = vqaddq_s64(v_sum_diff_total, fedcba98_76543210);
    }

    /* Update pointers for next iteration. */
    sig += sig_stride;
    mc_running_avg_y += mc_running_avg_y_stride;
    running_avg_y += running_avg_y_stride;
  }

  /* Too much adjustments => copy block. */
  {
    int64x1_t x = vqadd_s64(vget_high_s64(v_sum_diff_total),
                            vget_low_s64(v_sum_diff_total));
    int sum_diff = vget_lane_s32(vabs_s32(vreinterpret_s32_s64(x)), 0);
    int sum_diff_thresh = SUM_DIFF_THRESHOLD;

    if (increase_denoising) sum_diff_thresh = SUM_DIFF_THRESHOLD_HIGH;
    if (sum_diff > sum_diff_thresh) {
      // Before returning to copy the block (i.e., apply no denoising),
      // checK if we can still apply some (weaker) temporal filtering to
      // this block, that would otherwise not be denoised at all. Simplest
      // is to apply an additional adjustment to running_avg_y to bring it
      // closer to sig. The adjustment is capped by a maximum delta, and
      // chosen such that in most cases the resulting sum_diff will be
      // within the accceptable range given by sum_diff_thresh.

      // The delta is set by the excess of absolute pixel diff over the
      // threshold.
      int delta = ((sum_diff - sum_diff_thresh) >> 8) + 1;
      // Only apply the adjustment for max delta up to 3.
      if (delta < 4) {
        const uint8x16_t k_delta = vmovq_n_u8(delta);
        sig -= sig_stride * 16;
        mc_running_avg_y -= mc_running_avg_y_stride * 16;
        running_avg_y -= running_avg_y_stride * 16;
        for (r = 0; r < 16; ++r) {
          uint8x16_t v_running_avg_y = vld1q_u8(running_avg_y);
          const uint8x16_t v_sig = vld1q_u8(sig);
          const uint8x16_t v_mc_running_avg_y = vld1q_u8(mc_running_avg_y);

          /* Calculate absolute difference and sign masks. */
          const uint8x16_t v_abs_diff = vabdq_u8(v_sig, v_mc_running_avg_y);
          const uint8x16_t v_diff_pos_mask =
              vcltq_u8(v_sig, v_mc_running_avg_y);
          const uint8x16_t v_diff_neg_mask =
              vcgtq_u8(v_sig, v_mc_running_avg_y);
          // Clamp absolute difference to delta to get the adjustment.
          const uint8x16_t v_abs_adjustment = vminq_u8(v_abs_diff, (k_delta));

          const uint8x16_t v_pos_adjustment =
              vandq_u8(v_diff_pos_mask, v_abs_adjustment);
          const uint8x16_t v_neg_adjustment =
              vandq_u8(v_diff_neg_mask, v_abs_adjustment);

          v_running_avg_y = vqsubq_u8(v_running_avg_y, v_pos_adjustment);
          v_running_avg_y = vqaddq_u8(v_running_avg_y, v_neg_adjustment);

          /* Store results. */
          vst1q_u8(running_avg_y, v_running_avg_y);

          {
            const int8x16_t v_sum_diff =
                vqsubq_s8(vreinterpretq_s8_u8(v_neg_adjustment),
                          vreinterpretq_s8_u8(v_pos_adjustment));

            const int16x8_t fe_dc_ba_98_76_54_32_10 = vpaddlq_s8(v_sum_diff);
            const int32x4_t fedc_ba98_7654_3210 =
                vpaddlq_s16(fe_dc_ba_98_76_54_32_10);
            const int64x2_t fedcba98_76543210 =
                vpaddlq_s32(fedc_ba98_7654_3210);

            v_sum_diff_total = vqaddq_s64(v_sum_diff_total, fedcba98_76543210);
          }
          /* Update pointers for next iteration. */
          sig += sig_stride;
          mc_running_avg_y += mc_running_avg_y_stride;
          running_avg_y += running_avg_y_stride;
        }
        {
          // Update the sum of all pixel differences of this MB.
          x = vqadd_s64(vget_high_s64(v_sum_diff_total),
                        vget_low_s64(v_sum_diff_total));
          sum_diff = vget_lane_s32(vabs_s32(vreinterpret_s32_s64(x)), 0);

          if (sum_diff > sum_diff_thresh) {
            return COPY_BLOCK;
          }
        }
      } else {
        return COPY_BLOCK;
      }
    }
  }

  /* Tell above level that block was filtered. */
  running_avg_y -= running_avg_y_stride * 16;
  sig -= sig_stride * 16;

  vp8_copy_mem16x16(running_avg_y, running_avg_y_stride, sig, sig_stride);

  return FILTER_BLOCK;
}

int vp8_denoiser_filter_uv_neon(unsigned char *mc_running_avg,
                                int mc_running_avg_stride,
                                unsigned char *running_avg,
                                int running_avg_stride, unsigned char *sig,
                                int sig_stride, unsigned int motion_magnitude,
                                int increase_denoising) {
  /* If motion_magnitude is small, making the denoiser more aggressive by
   * increasing the adjustment for each level, level1 adjustment is
   * increased, the deltas stay the same.
   */
  int shift_inc =
      (increase_denoising && motion_magnitude <= MOTION_MAGNITUDE_THRESHOLD_UV)
          ? 1
          : 0;
  const uint8x16_t v_level1_adjustment = vmovq_n_u8(
      (motion_magnitude <= MOTION_MAGNITUDE_THRESHOLD_UV) ? 4 + shift_inc : 3);

  const uint8x16_t v_delta_level_1_and_2 = vdupq_n_u8(1);
  const uint8x16_t v_delta_level_2_and_3 = vdupq_n_u8(2);
  const uint8x16_t v_level1_threshold = vmovq_n_u8(4 + shift_inc);
  const uint8x16_t v_level2_threshold = vdupq_n_u8(8);
  const uint8x16_t v_level3_threshold = vdupq_n_u8(16);
  int64x2_t v_sum_diff_total = vdupq_n_s64(0);
  int r;

  {
    uint16x4_t v_sum_block = vdup_n_u16(0);

    // Avoid denoising color signal if its close to average level.
    for (r = 0; r < 8; ++r) {
      const uint8x8_t v_sig = vld1_u8(sig);
      const uint16x4_t _76_54_32_10 = vpaddl_u8(v_sig);
      v_sum_block = vqadd_u16(v_sum_block, _76_54_32_10);
      sig += sig_stride;
    }
    sig -= sig_stride * 8;
    {
      const uint32x2_t _7654_3210 = vpaddl_u16(v_sum_block);
      const uint64x1_t _76543210 = vpaddl_u32(_7654_3210);
      const int sum_block = vget_lane_s32(vreinterpret_s32_u64(_76543210), 0);
      if (abs(sum_block - (128 * 8 * 8)) < SUM_DIFF_FROM_AVG_THRESH_UV) {
        return COPY_BLOCK;
      }
    }
  }

  /* Go over lines. */
  for (r = 0; r < 4; ++r) {
    /* Load inputs. */
    const uint8x8_t v_sig_lo = vld1_u8(sig);
    const uint8x8_t v_sig_hi = vld1_u8(&sig[sig_stride]);
    const uint8x16_t v_sig = vcombine_u8(v_sig_lo, v_sig_hi);
    const uint8x8_t v_mc_running_avg_lo = vld1_u8(mc_running_avg);
    const uint8x8_t v_mc_running_avg_hi =
        vld1_u8(&mc_running_avg[mc_running_avg_stride]);
    const uint8x16_t v_mc_running_avg =
        vcombine_u8(v_mc_running_avg_lo, v_mc_running_avg_hi);
    /* Calculate absolute difference and sign masks. */
    const uint8x16_t v_abs_diff = vabdq_u8(v_sig, v_mc_running_avg);
    const uint8x16_t v_diff_pos_mask = vcltq_u8(v_sig, v_mc_running_avg);
    const uint8x16_t v_diff_neg_mask = vcgtq_u8(v_sig, v_mc_running_avg);

    /* Figure out which level that put us in. */
    const uint8x16_t v_level1_mask = vcleq_u8(v_level1_threshold, v_abs_diff);
    const uint8x16_t v_level2_mask = vcleq_u8(v_level2_threshold, v_abs_diff);
    const uint8x16_t v_level3_mask = vcleq_u8(v_level3_threshold, v_abs_diff);

    /* Calculate absolute adjustments for level 1, 2 and 3. */
    const uint8x16_t v_level2_adjustment =
        vandq_u8(v_level2_mask, v_delta_level_1_and_2);
    const uint8x16_t v_level3_adjustment =
        vandq_u8(v_level3_mask, v_delta_level_2_and_3);
    const uint8x16_t v_level1and2_adjustment =
        vaddq_u8(v_level1_adjustment, v_level2_adjustment);
    const uint8x16_t v_level1and2and3_adjustment =
        vaddq_u8(v_level1and2_adjustment, v_level3_adjustment);

    /* Figure adjustment absolute value by selecting between the absolute
     * difference if in level0 or the value for level 1, 2 and 3.
     */
    const uint8x16_t v_abs_adjustment =
        vbslq_u8(v_level1_mask, v_level1and2and3_adjustment, v_abs_diff);

    /* Calculate positive and negative adjustments. Apply them to the signal
     * and accumulate them. Adjustments are less than eight and the maximum
     * sum of them (7 * 16) can fit in a signed char.
     */
    const uint8x16_t v_pos_adjustment =
        vandq_u8(v_diff_pos_mask, v_abs_adjustment);
    const uint8x16_t v_neg_adjustment =
        vandq_u8(v_diff_neg_mask, v_abs_adjustment);

    uint8x16_t v_running_avg = vqaddq_u8(v_sig, v_pos_adjustment);
    v_running_avg = vqsubq_u8(v_running_avg, v_neg_adjustment);

    /* Store results. */
    vst1_u8(running_avg, vget_low_u8(v_running_avg));
    vst1_u8(&running_avg[running_avg_stride], vget_high_u8(v_running_avg));

    /* Sum all the accumulators to have the sum of all pixel differences
     * for this macroblock.
     */
    {
      const int8x16_t v_sum_diff =
          vqsubq_s8(vreinterpretq_s8_u8(v_pos_adjustment),
                    vreinterpretq_s8_u8(v_neg_adjustment));

      const int16x8_t fe_dc_ba_98_76_54_32_10 = vpaddlq_s8(v_sum_diff);

      const int32x4_t fedc_ba98_7654_3210 =
          vpaddlq_s16(fe_dc_ba_98_76_54_32_10);

      const int64x2_t fedcba98_76543210 = vpaddlq_s32(fedc_ba98_7654_3210);

      v_sum_diff_total = vqaddq_s64(v_sum_diff_total, fedcba98_76543210);
    }

    /* Update pointers for next iteration. */
    sig += sig_stride * 2;
    mc_running_avg += mc_running_avg_stride * 2;
    running_avg += running_avg_stride * 2;
  }

  /* Too much adjustments => copy block. */
  {
    int64x1_t x = vqadd_s64(vget_high_s64(v_sum_diff_total),
                            vget_low_s64(v_sum_diff_total));
    int sum_diff = vget_lane_s32(vabs_s32(vreinterpret_s32_s64(x)), 0);
    int sum_diff_thresh = SUM_DIFF_THRESHOLD_UV;
    if (increase_denoising) sum_diff_thresh = SUM_DIFF_THRESHOLD_HIGH_UV;
    if (sum_diff > sum_diff_thresh) {
      // Before returning to copy the block (i.e., apply no denoising),
      // checK if we can still apply some (weaker) temporal filtering to
      // this block, that would otherwise not be denoised at all. Simplest
      // is to apply an additional adjustment to running_avg_y to bring it
      // closer to sig. The adjustment is capped by a maximum delta, and
      // chosen such that in most cases the resulting sum_diff will be
      // within the accceptable range given by sum_diff_thresh.

      // The delta is set by the excess of absolute pixel diff over the
      // threshold.
      int delta = ((sum_diff - sum_diff_thresh) >> 8) + 1;
      // Only apply the adjustment for max delta up to 3.
      if (delta < 4) {
        const uint8x16_t k_delta = vmovq_n_u8(delta);
        sig -= sig_stride * 8;
        mc_running_avg -= mc_running_avg_stride * 8;
        running_avg -= running_avg_stride * 8;
        for (r = 0; r < 4; ++r) {
          const uint8x8_t v_sig_lo = vld1_u8(sig);
          const uint8x8_t v_sig_hi = vld1_u8(&sig[sig_stride]);
          const uint8x16_t v_sig = vcombine_u8(v_sig_lo, v_sig_hi);
          const uint8x8_t v_mc_running_avg_lo = vld1_u8(mc_running_avg);
          const uint8x8_t v_mc_running_avg_hi =
              vld1_u8(&mc_running_avg[mc_running_avg_stride]);
          const uint8x16_t v_mc_running_avg =
              vcombine_u8(v_mc_running_avg_lo, v_mc_running_avg_hi);
          /* Calculate absolute difference and sign masks. */
          const uint8x16_t v_abs_diff = vabdq_u8(v_sig, v_mc_running_avg);
          const uint8x16_t v_diff_pos_mask = vcltq_u8(v_sig, v_mc_running_avg);
          const uint8x16_t v_diff_neg_mask = vcgtq_u8(v_sig, v_mc_running_avg);
          // Clamp absolute difference to delta to get the adjustment.
          const uint8x16_t v_abs_adjustment = vminq_u8(v_abs_diff, (k_delta));

          const uint8x16_t v_pos_adjustment =
              vandq_u8(v_diff_pos_mask, v_abs_adjustment);
          const uint8x16_t v_neg_adjustment =
              vandq_u8(v_diff_neg_mask, v_abs_adjustment);
          const uint8x8_t v_running_avg_lo = vld1_u8(running_avg);
          const uint8x8_t v_running_avg_hi =
              vld1_u8(&running_avg[running_avg_stride]);
          uint8x16_t v_running_avg =
              vcombine_u8(v_running_avg_lo, v_running_avg_hi);

          v_running_avg = vqsubq_u8(v_running_avg, v_pos_adjustment);
          v_running_avg = vqaddq_u8(v_running_avg, v_neg_adjustment);

          /* Store results. */
          vst1_u8(running_avg, vget_low_u8(v_running_avg));
          vst1_u8(&running_avg[running_avg_stride],
                  vget_high_u8(v_running_avg));

          {
            const int8x16_t v_sum_diff =
                vqsubq_s8(vreinterpretq_s8_u8(v_neg_adjustment),
                          vreinterpretq_s8_u8(v_pos_adjustment));

            const int16x8_t fe_dc_ba_98_76_54_32_10 = vpaddlq_s8(v_sum_diff);
            const int32x4_t fedc_ba98_7654_3210 =
                vpaddlq_s16(fe_dc_ba_98_76_54_32_10);
            const int64x2_t fedcba98_76543210 =
                vpaddlq_s32(fedc_ba98_7654_3210);

            v_sum_diff_total = vqaddq_s64(v_sum_diff_total, fedcba98_76543210);
          }
          /* Update pointers for next iteration. */
          sig += sig_stride * 2;
          mc_running_avg += mc_running_avg_stride * 2;
          running_avg += running_avg_stride * 2;
        }
        {
          // Update the sum of all pixel differences of this MB.
          x = vqadd_s64(vget_high_s64(v_sum_diff_total),
                        vget_low_s64(v_sum_diff_total));
          sum_diff = vget_lane_s32(vabs_s32(vreinterpret_s32_s64(x)), 0);

          if (sum_diff > sum_diff_thresh) {
            return COPY_BLOCK;
          }
        }
      } else {
        return COPY_BLOCK;
      }
    }
  }

  /* Tell above level that block was filtered. */
  running_avg -= running_avg_stride * 8;
  sig -= sig_stride * 8;

  vp8_copy_mem8x8(running_avg, running_avg_stride, sig, sig_stride);

  return FILTER_BLOCK;
}