Improve near lossless compression when a prediction filter is used.

The old implementation in enc/near_lossless.c performing a separate
preprocessing step is used only when a prediction filter is not used,
otherwise a new implementation integrated into lossless_enc.c is used.

It retains the same logic for converting near lossless quality into max
number of bits dropped, and for adjusting the number of bits based on
the smoothness of the image at a given pixel. As before, borders are not
changed.

Then, instead of quantizing raw component values, the residual after
subtract green and after prediction is quantized according to the
resulting number of bits, taking care to not cross the boundary between
255 and 0 after decoding. Ties are resolved by moving closer to the
prediction instead of by bankers’ rounding.

This results in about 15% size decrease for the same quality.

Change-Id: If3e9c388158c2e3e75ef88876703f40b932f671f

src/dsp/lossless.h

@@ -158,7 +158,8 @@ void VP8LCollectColorBlueTransforms_C(const uint32_t* argb, int stride,
 
 void VP8LResidualImage(int width, int height, int bits, int low_effort,
                        uint32_t* const argb, uint32_t* const argb_scratch,
-                       uint32_t* const image, int exact);
+                       uint32_t* const image, int near_lossless, int exact,
+                       int used_subtract_green);
 
 void VP8LColorSpaceTransform(int width, int height, int bits, int quality,
                              uint32_t* const argb, uint32_t* image);
@@ -172,6 +173,17 @@ static WEBP_INLINE uint32_t VP8LSubSampleSize(uint32_t size,
   return (size + (1 << sampling_bits) - 1) >> sampling_bits;
 }
 
+// Converts near lossless quality into max number of bits shaved off.
+static WEBP_INLINE int VP8LNearLosslessBits(int near_lossless_quality) {
+  //    100 -> 0
+  // 80..99 -> 1
+  // 60..79 -> 2
+  // 40..59 -> 3
+  // 20..39 -> 4
+  //  0..19 -> 5
+  return 5 - near_lossless_quality / 20;
+}
+
 // -----------------------------------------------------------------------------
 // Faster logarithm for integers. Small values use a look-up table.
 
@@ -336,7 +348,7 @@ static WEBP_INLINE uint32_t VP8LAddPixels(uint32_t a, uint32_t b) {
   return (alpha_and_green & 0xff00ff00u) | (red_and_blue & 0x00ff00ffu);
 }
 
-// Difference of each component with mod 256.
+// Difference of each component, mod 256.
 static WEBP_INLINE uint32_t VP8LSubPixels(uint32_t a, uint32_t b) {
   const uint32_t alpha_and_green =
       0x00ff00ffu + (a & 0xff00ff00u) - (b & 0xff00ff00u);

src/dsp/lossless_enc.c

@@ -382,6 +382,7 @@ static float FastLog2Slow(uint32_t v) {
 
 // Mostly used to reduce code size + readability
 static WEBP_INLINE int GetMin(int a, int b) { return (a > b) ? b : a; }
+static WEBP_INLINE int GetMax(int a, int b) { return (a < b) ? b : a; }
 
 //------------------------------------------------------------------------------
 // Methods to calculate Entropy (Shannon).
@@ -551,18 +552,205 @@ static WEBP_INLINE uint32_t Predict(VP8LPredictorFunc pred_func,
   }
 }
 
+static int MaxDiffBetweenPixels(uint32_t p1, uint32_t p2) {
+  const int diff_a = abs((int)(p1 >> 24) - (int)(p2 >> 24));
+  const int diff_r = abs((int)((p1 >> 16) & 0xff) - (int)((p2 >> 16) & 0xff));
+  const int diff_g = abs((int)((p1 >> 8) & 0xff) - (int)((p2 >> 8) & 0xff));
+  const int diff_b = abs((int)(p1 & 0xff) - (int)(p2 & 0xff));
+  return GetMax(GetMax(diff_a, diff_r), GetMax(diff_g, diff_b));
+}
+
+static int MaxDiffAroundPixel(uint32_t current, uint32_t up, uint32_t down,
+                              uint32_t left, uint32_t right) {
+  const int diff_up = MaxDiffBetweenPixels(current, up);
+  const int diff_down = MaxDiffBetweenPixels(current, down);
+  const int diff_left = MaxDiffBetweenPixels(current, left);
+  const int diff_right = MaxDiffBetweenPixels(current, right);
+  return GetMax(GetMax(diff_up, diff_down), GetMax(diff_left, diff_right));
+}
+
+static uint32_t AddGreenToBlueAndRed(uint32_t argb) {
+  const uint32_t green = (argb >> 8) & 0xff;
+  uint32_t red_blue = argb & 0x00ff00ffu;
+  red_blue += (green << 16) | green;
+  red_blue &= 0x00ff00ffu;
+  return (argb & 0xff00ff00u) | red_blue;
+}
+
+static void MaxDiffsForRow(int width, int stride, const uint32_t* const argb,
+                           uint8_t* const max_diffs, int used_subtract_green) {
+  uint32_t current, up, down, left, right;
+  int x;
+  if (width <= 2) return;
+  current = argb[0];
+  right = argb[1];
+  if (used_subtract_green) {
+    current = AddGreenToBlueAndRed(current);
+    right = AddGreenToBlueAndRed(right);
+  }
+  // max_diffs[0] and max_diffs[width - 1] are never used.
+  for (x = 1; x < width - 1; ++x) {
+    up = argb[-stride + x];
+    down = argb[stride + x];
+    left = current;
+    current = right;
+    right = argb[x + 1];
+    if (used_subtract_green) {
+      up = AddGreenToBlueAndRed(up);
+      down = AddGreenToBlueAndRed(down);
+      right = AddGreenToBlueAndRed(right);
+    }
+    max_diffs[x] = MaxDiffAroundPixel(current, up, down, left, right);
+  }
+}
+
+// Quantize the difference between the actual component value and its prediction
+// to a multiple of quantization, working modulo 256, taking care not to cross
+// a boundary (inclusive upper limit).
+static uint8_t NearLosslessComponent(uint8_t value, uint8_t predict,
+                                     uint8_t boundary, int quantization) {
+  const int residual = (value - predict) & 0xff;
+  const int boundary_residual = (boundary - predict) & 0xff;
+  const int lower = residual & ~(quantization - 1);
+  const int upper = lower + quantization;
+  // Resolve ties towards a value closer to the prediction (i.e. towards lower
+  // if value comes after prediction and towards upper otherwise).
+  const int bias = ((boundary - value) & 0xff) < boundary_residual;
+  if (residual - lower < upper - residual + bias) {
+    // lower is closer to residual than upper.
+    if (residual > boundary_residual && lower <= boundary_residual) {
+      // Halve quantization step to avoid crossing boundary. This midpoint is
+      // on the same side of boundary as residual because midpoint >= residual
+      // (since lower is closer than upper) and residual is above the boundary.
+      return lower + (quantization >> 1);
+    }
+    return lower;
+  } else {
+    // upper is closer to residual than lower.
+    if (residual <= boundary_residual && upper > boundary_residual) {
+      // Halve quantization step to avoid crossing boundary. This midpoint is
+      // on the same side of boundary as residual because midpoint <= residual
+      // (since upper is closer than lower) and residual is below the boundary.
+      return lower + (quantization >> 1);
+    }
+    return upper & 0xff;
+  }
+}
+
+// Quantize every component of the difference between the actual pixel value and
+// its prediction to a multiple of a quantization (a power of 2, not larger than
+// max_quantization which is a power of 2, smaller than max_diff). Take care if
+// value and predict have undergone subtract green, which means that red and
+// blue are represented as offsets from green.
+static uint32_t NearLossless(uint32_t value, uint32_t predict,
+                             int max_quantization, int max_diff,
+                             int used_subtract_green) {
+  int quantization;
+  uint8_t new_green = 0;
+  uint8_t green_diff = 0;
+  uint8_t a, r, g, b;
+  if (max_diff <= 2) {
+    return VP8LSubPixels(value, predict);
+  }
+  quantization = max_quantization;
+  while (quantization >= max_diff) {
+    quantization >>= 1;
+  }
+  if ((value >> 24) == 0 || (value >> 24) == 0xff) {
+    // Preserve transparency of fully transparent or fully opaque pixels.
+    a = ((value >> 24) - (predict >> 24)) & 0xff;
+  } else {
+    a = NearLosslessComponent(value >> 24, predict >> 24, 0xff, quantization);
+  }
+  g = NearLosslessComponent((value >> 8) & 0xff, (predict >> 8) & 0xff, 0xff,
+                            quantization);
+  if (used_subtract_green) {
+    // The green offset will be added to red and blue components during decoding
+    // to obtain the actual red and blue values.
+    new_green = ((predict >> 8) + g) & 0xff;
+    // The amount by which green has been adjusted during quantization. It is
+    // subtracted from red and blue for compensation, to avoid accumulating two
+    // quantization errors in them.
+    green_diff = (new_green - (value >> 8)) & 0xff;
+  }
+  r = NearLosslessComponent(((value >> 16) - green_diff) & 0xff,
+                            (predict >> 16) & 0xff, 0xff - new_green,
+                            quantization);
+  b = NearLosslessComponent((value - green_diff) & 0xff, predict & 0xff,
+                            0xff - new_green, quantization);
+  return ((uint32_t)a << 24) | ((uint32_t)r << 16) | ((uint32_t)g << 8) | b;
+}
+
+// Returns the difference between the pixel and its prediction. In case of a
+// lossy encoding, updates the source image to avoid propagating the deviation
+// further to pixels which depend on the current pixel for their predictions.
+static uint32_t GetResidual(int width, int height,
+                            uint32_t* const upper_row,
+                            uint32_t* const current_row,
+                            const uint8_t* const max_diffs,
+                            int mode, VP8LPredictorFunc pred_func,
+                            int x, int y,
+                            int max_quantization,
+                            int exact, int used_subtract_green) {
+  const uint32_t predict = Predict(pred_func, x, y, current_row, upper_row);
+  uint32_t residual;
+  if (max_quantization == 1 || mode == 0 || y == 0 || y == height - 1 ||
+      x == 0 || x == width - 1) {
+    residual = VP8LSubPixels(current_row[x], predict);
+  } else {
+    residual = NearLossless(current_row[x], predict, max_quantization,
+                            max_diffs[x], used_subtract_green);
+    // Update the source image.
+    current_row[x] = VP8LAddPixels(predict, residual);
+    // x is never 0 here so we do not need to update upper_row like below.
+  }
+  if (!exact && (current_row[x] & kMaskAlpha) == 0) {
+    // If alpha is 0, cleanup RGB. We can choose the RGB values of the residual
+    // for best compression. The prediction of alpha itself can be non-zero and
+    // must be kept though. We choose RGB of the residual to be 0.
+    residual &= kMaskAlpha;
+    // Update the source image.
+    current_row[x] = predict & ~kMaskAlpha;
+    // The prediction for the rightmost pixel in a row uses the leftmost pixel
+    // in that row as its top-right context pixel. Hence if we change the
+    // leftmost pixel of current_row, the corresponding change must be applied
+    // to upper_row as well where top-right context is being read from.
+    if (x == 0 && y != 0) upper_row[width] = current_row[0];
+  }
+  return residual;
+}
+
 // Returns best predictor and updates the accumulated histogram.
+// If max_quantization > 1, assumes that near lossless processing will be
+// applied, quantizing residuals to multiples of quantization levels up to
+// max_quantization (the actual quantization level depends on smoothness near
+// the given pixel).
 static int GetBestPredictorForTile(int width, int height,
                                    int tile_x, int tile_y, int bits,
                                    int accumulated[4][256],
-                                   const uint32_t* const argb_scratch,
+                                   uint32_t* const argb_scratch,
-                                   int exact) {
+                                   const uint32_t* const argb,
+                                   int max_quantization,
+                                   int exact, int used_subtract_green) {
   const int kNumPredModes = 14;
-  const int col_start = tile_x << bits;
+  const int start_x = tile_x << bits;
-  const int row_start = tile_y << bits;
+  const int start_y = tile_y << bits;
   const int tile_size = 1 << bits;
-  const int max_y = GetMin(tile_size, height - row_start);
+  const int max_y = GetMin(tile_size, height - start_y);
-  const int max_x = GetMin(tile_size, width - col_start);
+  const int max_x = GetMin(tile_size, width - start_x);
+  // Whether there exist columns just outside the tile.
+  const int have_left = (start_x > 0);
+  const int have_right = (max_x < width - start_x);
+  // Position and size of the strip covering the tile and adjacent columns if
+  // they exist.
+  const int context_start_x = start_x - have_left;
+  const int context_width = max_x + have_left + have_right;
+  // The width of upper_row and current_row is one pixel larger than image width
+  // to allow the top right pixel to point to the leftmost pixel of the next row
+  // when at the right edge.
+  uint32_t* upper_row = argb_scratch;
+  uint32_t* current_row = upper_row + width + 1;
+  uint8_t* const max_diffs = (uint8_t*)(current_row + width + 1);
   float best_diff = MAX_DIFF_COST;
   int best_mode = 0;
   int mode;
@@ -571,28 +759,46 @@ static int GetBestPredictorForTile(int width, int height,
   // Need pointers to be able to swap arrays.
   int (*histo_argb)[256] = histo_stack_1;
   int (*best_histo)[256] = histo_stack_2;
-
   int i, j;
+
   for (mode = 0; mode < kNumPredModes; ++mode) {
-    const uint32_t* current_row = argb_scratch;
     const VP8LPredictorFunc pred_func = VP8LPredictors[mode];
     float cur_diff;
-    int y;
+    int relative_y;
     memset(histo_argb, 0, sizeof(histo_stack_1));
-    for (y = 0; y < max_y; ++y) {
+    if (start_y > 0) {
-      int x;
+      // Read the row above the tile which will become the first upper_row.
-      const int row = row_start + y;
+      // Include a pixel to the left if it exists; include a pixel to the right
-      const uint32_t* const upper_row = current_row;
+      // in all cases (wrapping to the leftmost pixel of the next row if it does
-      current_row = upper_row + width;
+      // not exist).
-      for (x = 0; x < max_x; ++x) {
+      memcpy(current_row + context_start_x,
-        const int col = col_start + x;
+             argb + (start_y - 1) * width + context_start_x,
-        const uint32_t predict =
+             sizeof(*argb) * (max_x + have_left + 1));
-            Predict(pred_func, col, row, current_row, upper_row);
+    }
-        uint32_t residual = VP8LSubPixels(current_row[col], predict);
+    for (relative_y = 0; relative_y < max_y; ++relative_y) {
-        if (!exact && (current_row[col] & kMaskAlpha) == 0) {
+      const int y = start_y + relative_y;
-          residual &= kMaskAlpha;  // See CopyTileWithPrediction.
+      int relative_x;
-        }
+      uint32_t* tmp = upper_row;
-        UpdateHisto(histo_argb, residual);
+      upper_row = current_row;
+      current_row = tmp;
+      // Read current_row. Include a pixel to the left if it exists; include a
+      // pixel to the right in all cases except at the bottom right corner of
+      // the image (wrapping to the leftmost pixel of the next row if it does
+      // not exist in the current row).
+      memcpy(current_row + context_start_x,
+             argb + y * width + context_start_x,
+             sizeof(*argb) * (max_x + have_left + (y + 1 < height)));
+      if (max_quantization > 1 && y >= 1 && y + 1 < height) {
+        MaxDiffsForRow(context_width, width, argb + y * width + context_start_x,
+                       max_diffs + context_start_x, used_subtract_green);
+      }
+
+      for (relative_x = 0; relative_x < max_x; ++relative_x) {
+        const int x = start_x + relative_x;
+        UpdateHisto(histo_argb,
+                    GetResidual(width, height, upper_row, current_row,
+                                max_diffs, mode, pred_func, x, y,
+                                max_quantization, exact, used_subtract_green));
       }
     }
     cur_diff = PredictionCostSpatialHistogram(
@@ -615,71 +821,82 @@ static int GetBestPredictorForTile(int width, int height,
   return best_mode;
 }
 
+// Converts pixels of the image to residuals with respect to predictions.
+// If max_quantization > 1, applies near lossless processing, quantizing
+// residuals to multiples of quantization levels up to max_quantization
+// (the actual quantization level depends on smoothness near the given pixel).
 static void CopyImageWithPrediction(int width, int height,
                                     int bits, uint32_t* const modes,
                                     uint32_t* const argb_scratch,
                                     uint32_t* const argb,
-                                    int low_effort, int exact) {
+                                    int low_effort, int max_quantization,
+                                    int exact, int used_subtract_green) {
   const int tiles_per_row = VP8LSubSampleSize(width, bits);
   const int mask = (1 << bits) - 1;
-  // The row size is one pixel longer to allow the top right pixel to point to
+  // The width of upper_row and current_row is one pixel larger than image width
-  // the leftmost pixel of the next row when at the right edge.
+  // to allow the top right pixel to point to the leftmost pixel of the next row
-  uint32_t* current_row = argb_scratch;
+  // when at the right edge.
-  uint32_t* upper_row = argb_scratch + width + 1;
+  uint32_t* upper_row = argb_scratch;
+  uint32_t* current_row = upper_row + width + 1;
+  uint8_t* current_max_diffs = (uint8_t*)(current_row + width + 1);
+  uint8_t* lower_max_diffs = current_max_diffs + width;
   int y;
-  VP8LPredictorFunc pred_func =
+  int mode = 0;
-      low_effort ? VP8LPredictors[kPredLowEffort] : NULL;
+  VP8LPredictorFunc pred_func = NULL;
 
   for (y = 0; y < height; ++y) {
     int x;
-    uint32_t* tmp = upper_row;
+    uint32_t* const tmp32 = upper_row;
     upper_row = current_row;
-    current_row = tmp;
+    current_row = tmp32;
-    memcpy(current_row, argb + y * width, sizeof(*current_row) * width);
+    memcpy(current_row, argb + y * width,
-    current_row[width] = (y + 1 < height) ? argb[(y + 1) * width] : ARGB_BLACK;
+           sizeof(*argb) * (width + (y + 1 < height)));
 
     if (low_effort) {
       for (x = 0; x < width; ++x) {
-        const uint32_t predict =
+        const uint32_t predict = Predict(VP8LPredictors[kPredLowEffort], x, y,
-            Predict(pred_func, x, y, current_row, upper_row);
+                                         current_row, upper_row);
         argb[y * width + x] = VP8LSubPixels(current_row[x], predict);
       }
     } else {
+      if (max_quantization > 1) {
+        // Compute max_diffs for the lower row now, because that needs the
+        // contents of argb for the current row, which we will overwrite with
+        // residuals before proceeding with the next row.
+        uint8_t* const tmp8 = current_max_diffs;
+        current_max_diffs = lower_max_diffs;
+        lower_max_diffs = tmp8;
+        if (y + 2 < height) {
+          MaxDiffsForRow(width, width, argb + (y + 1) * width, lower_max_diffs,
+                         used_subtract_green);
+        }
+      }
       for (x = 0; x < width; ++x) {
-        uint32_t predict, residual;
         if ((x & mask) == 0) {
-          const int mode =
+          mode = (modes[(y >> bits) * tiles_per_row + (x >> bits)] >> 8) & 0xff;
-              (modes[(y >> bits) * tiles_per_row + (x >> bits)] >> 8) & 0xff;
           pred_func = VP8LPredictors[mode];
         }
-        predict = Predict(pred_func, x, y, current_row, upper_row);
+        argb[y * width + x] = GetResidual(
-        residual = VP8LSubPixels(current_row[x], predict);
+            width, height, upper_row, current_row, current_max_diffs, mode,
-        if (!exact && (current_row[x] & kMaskAlpha) == 0) {
+            pred_func, x, y, max_quantization, exact, used_subtract_green);
-          // If alpha is 0, cleanup RGB. We can choose the RGB values of the
-          // residual for best compression. The prediction of alpha itself can
-          // be non-zero and must be kept though. We choose RGB of the residual
-          // to be 0.
-          residual &= kMaskAlpha;
-          // Update input image so that next predictions use correct RGB value.
-          current_row[x] = predict & ~kMaskAlpha;
-          if (x == 0 && y != 0) upper_row[width] = current_row[x];
-        }
-        argb[y * width + x] = residual;
       }
     }
   }
 }
 
+// Finds the best predictor for each tile, and converts the image to residuals
+// with respect to predictions. If near_lossless_quality < 100, applies
+// near lossless processing, shaving off more bits of residuals for lower
+// qualities.
 void VP8LResidualImage(int width, int height, int bits, int low_effort,
                        uint32_t* const argb, uint32_t* const argb_scratch,
-                       uint32_t* const image, int exact) {
+                       uint32_t* const image, int near_lossless_quality,
-  const int max_tile_size = 1 << bits;
+                       int exact, int used_subtract_green) {
   const int tiles_per_row = VP8LSubSampleSize(width, bits);
   const int tiles_per_col = VP8LSubSampleSize(height, bits);
-  uint32_t* const upper_row = argb_scratch;
-  uint32_t* const current_tile_rows = argb_scratch + width;
   int tile_y;
   int histo[4][256];
+  const int max_quantization = 1 << VP8LNearLosslessBits(near_lossless_quality);
   if (low_effort) {
     int i;
     for (i = 0; i < tiles_per_row * tiles_per_col; ++i) {
@@ -688,26 +905,19 @@ void VP8LResidualImage(int width, int height, int bits, int low_effort,
   } else {
     memset(histo, 0, sizeof(histo));
     for (tile_y = 0; tile_y < tiles_per_col; ++tile_y) {
-      const int tile_y_offset = tile_y * max_tile_size;
-      const int this_tile_height =
-          (tile_y < tiles_per_col - 1) ? max_tile_size : height - tile_y_offset;
       int tile_x;
-      if (tile_y > 0) {
-        memcpy(upper_row, current_tile_rows + (max_tile_size - 1) * width,
-               width * sizeof(*upper_row));
-      }
-      memcpy(current_tile_rows, &argb[tile_y_offset * width],
-             this_tile_height * width * sizeof(*current_tile_rows));
       for (tile_x = 0; tile_x < tiles_per_row; ++tile_x) {
         const int pred = GetBestPredictorForTile(width, height, tile_x, tile_y,
-            bits, (int (*)[256])histo, argb_scratch, exact);
+            bits, histo, argb_scratch, argb, max_quantization, exact,
+            used_subtract_green);
         image[tile_y * tiles_per_row + tile_x] = ARGB_BLACK | (pred << 8);
       }
     }
   }
 
-  CopyImageWithPrediction(width, height, bits,
+  CopyImageWithPrediction(width, height, bits, image, argb_scratch, argb,
-                          image, argb_scratch, argb, low_effort, exact);
+                          low_effort, max_quantization, exact,
+                          used_subtract_green);
 }
 
 void VP8LSubtractGreenFromBlueAndRed_C(uint32_t* argb_data, int num_pixels) {

src/enc/near_lossless.c

@@ -97,22 +97,11 @@ static void NearLossless(int xsize, int ysize, uint32_t* argb,
   }
 }
 
-static int QualityToLimitBits(int quality) {
-  // quality mapping:
-  //   0..19 -> 5
-  //  20..39 -> 4
-  //  40..59 -> 3
-  //  60..79 -> 2
-  //  80..99 -> 1
-  //  100    -> 0
-  return MAX_LIMIT_BITS - quality / 20;
-}
-
 int VP8ApplyNearLossless(int xsize, int ysize, uint32_t* argb, int quality) {
   int i;
   uint32_t* const copy_buffer =
       (uint32_t*)WebPSafeMalloc(xsize * 3, sizeof(*copy_buffer));
-  const int limit_bits = QualityToLimitBits(quality);
+  const int limit_bits = VP8LNearLosslessBits(quality);
   assert(argb != NULL);
   assert(limit_bits >= 0);
   assert(limit_bits <= MAX_LIMIT_BITS);

src/enc/vp8l.c

@@ -1008,6 +1008,7 @@ static void ApplySubtractGreen(VP8LEncoder* const enc, int width, int height,
 static WebPEncodingError ApplyPredictFilter(const VP8LEncoder* const enc,
                                             int width, int height,
                                             int quality, int low_effort,
+                                            int used_subtract_green,
                                             VP8LBitWriter* const bw) {
   const int pred_bits = enc->transform_bits_;
   const int transform_width = VP8LSubSampleSize(width, pred_bits);
@@ -1015,7 +1016,8 @@ static WebPEncodingError ApplyPredictFilter(const VP8LEncoder* const enc,
 
   VP8LResidualImage(width, height, pred_bits, low_effort, enc->argb_,
                     enc->argb_scratch_, enc->transform_data_,
-                    enc->config_->exact);
+                    enc->config_->near_lossless, enc->config_->exact,
+                    used_subtract_green);
   VP8LPutBits(bw, TRANSFORM_PRESENT, 1);
   VP8LPutBits(bw, PREDICTOR_TRANSFORM, 2);
   assert(pred_bits >= 2);
@@ -1129,21 +1131,26 @@ static void ClearTransformBuffer(VP8LEncoder* const enc) {
 static WebPEncodingError AllocateTransformBuffer(VP8LEncoder* const enc,
                                                  int width, int height) {
   WebPEncodingError err = VP8_ENC_OK;
-  const int tile_size = 1 << enc->transform_bits_;
   const uint64_t image_size = width * height;
-  // Ensure enough size for tiles, as well as for two scanlines and two
+  // VP8LResidualImage needs room for 2 scanlines of uint32 pixels with an extra
-  // extra pixels for CopyImageWithPrediction.
+  // pixel in each, plus 2 regular scanlines of bytes.
+  // TODO(skal): Clean up by using arithmetic in bytes instead of words.
   const uint64_t argb_scratch_size =
-      enc->use_predict_ ? tile_size * width + width + 2 : 0;
+      enc->use_predict_
-  const int transform_data_size =
+          ? (width + 1) * 2 +
+            (width * 2 + sizeof(uint32_t) - 1) / sizeof(uint32_t)
+          : 0;
+  const uint64_t transform_data_size =
       (enc->use_predict_ || enc->use_cross_color_)
           ? VP8LSubSampleSize(width, enc->transform_bits_) *
-            VP8LSubSampleSize(height, enc->transform_bits_)
+                VP8LSubSampleSize(height, enc->transform_bits_)
           : 0;
+  const uint64_t max_alignment_in_words =
+      (WEBP_ALIGN_CST + sizeof(uint32_t) - 1) / sizeof(uint32_t);
   const uint64_t mem_size =
-      image_size + WEBP_ALIGN_CST +
+      image_size + max_alignment_in_words +
-      argb_scratch_size + WEBP_ALIGN_CST +
+      argb_scratch_size + max_alignment_in_words +
-      (uint64_t)transform_data_size;
+      transform_data_size;
   uint32_t* mem = enc->transform_mem_;
   if (mem == NULL || mem_size > enc->transform_mem_size_) {
     ClearTransformBuffer(enc);
@@ -1412,7 +1419,8 @@ WebPEncodingError VP8LEncodeStream(const WebPConfig* const config,
   }
 
   // Apply near-lossless preprocessing.
-  use_near_lossless = !enc->use_palette_ && (config->near_lossless < 100);
+  use_near_lossless =
+      (config->near_lossless < 100) && !enc->use_palette_ && !enc->use_predict_;
   if (use_near_lossless) {
     if (!VP8ApplyNearLossless(width, height, picture->argb,
                               config->near_lossless)) {
@@ -1464,7 +1472,7 @@ WebPEncodingError VP8LEncodeStream(const WebPConfig* const config,
 
     if (enc->use_predict_) {
       err = ApplyPredictFilter(enc, enc->current_width_, height, quality,
-                               low_effort, bw);
+                               low_effort, enc->use_subtract_green_, bw);
       if (err != VP8_ENC_OK) goto Error;
     }