h264: fully support cropping.

Based on a patch by Vittorio Giovara <vittorio.giovara@gmail.com>

Fixes Bug 378.

6 files changed, 105 insertions, 39 deletions

diff --git a/libavcodec/avcodec.h b/libavcodec/avcodec.h
index 84a685986b..c36d976a52 100644
--- a/libavcodec/avcodec.h
+++ b/libavcodec/avcodec.h
@@ -609,6 +609,11 @@ typedef struct RcOverride{
    Note: Not everything is supported yet.
 */
 
+/**
+ * Allow decoders to produce frames with data planes that are not aligned
+ * to CPU requirements (e.g. due to cropping).
+ */
+#define CODEC_FLAG_UNALIGNED 0x0001
 #define CODEC_FLAG_QSCALE 0x0002  ///< Use fixed qscale.
 #define CODEC_FLAG_4MV    0x0004  ///< 4 MV per MB allowed / advanced prediction for H.263.
 #define CODEC_FLAG_QPEL   0x0010  ///< Use qpel MC.
diff --git a/libavcodec/h264.c b/libavcodec/h264.c
index 4a3d2e7512..d8bc824d5c 100644
--- a/libavcodec/h264.c
+++ b/libavcodec/h264.c
@@ -1426,9 +1426,6 @@ av_cold int ff_h264_decode_init(AVCodecContext *avctx)
 
     h->avctx = avctx;
 
-    h->width    = h->avctx->width;
-    h->height   = h->avctx->height;
-
     h->bit_depth_luma    = 8;
     h->chroma_format_idc = 1;
 
@@ -2978,14 +2975,49 @@ static enum AVPixelFormat get_pixel_format(H264Context *h)
     }
 }
 
+/* export coded and cropped frame dimensions to AVCodecContext */
+static int init_dimensions(H264Context *h)
+{
+    int width  = h->width  - (h->sps.crop_right + h->sps.crop_left);
+    int height = h->height - (h->sps.crop_top   + h->sps.crop_bottom);
+
+    /* handle container cropping */
+    if (!h->sps.crop &&
+        FFALIGN(h->avctx->width,  16) == h->width &&
+        FFALIGN(h->avctx->height, 16) == h->height) {
+        width  = h->avctx->width;
+        height = h->avctx->height;
+    }
+
+    if (width <= 0 || height <= 0) {
+        av_log(h->avctx, AV_LOG_ERROR, "Invalid cropped dimensions: %dx%d.\n",
+               width, height);
+        if (h->avctx->err_recognition & AV_EF_EXPLODE)
+            return AVERROR_INVALIDDATA;
+
+        av_log(h->avctx, AV_LOG_WARNING, "Ignoring cropping information.\n");
+        h->sps.crop_bottom = h->sps.crop_top = h->sps.crop_right = h->sps.crop_left = 0;
+        h->sps.crop = 0;
+
+        width  = h->width;
+        height = h->height;
+    }
+
+    h->avctx->coded_width  = h->width;
+    h->avctx->coded_height = h->height;
+    h->avctx->width        = width;
+    h->avctx->height       = height;
+
+    return 0;
+}
+
 static int h264_slice_header_init(H264Context *h, int reinit)
 {
     int nb_slices = (HAVE_THREADS &&
                      h->avctx->active_thread_type & FF_THREAD_SLICE) ?
                     h->avctx->thread_count : 1;
-    int i;
+    int i, ret;
 
-    avcodec_set_dimensions(h->avctx, h->width, h->height);
     h->avctx->sample_aspect_ratio = h->sps.sar;
     av_assert0(h->avctx->sample_aspect_ratio.den);
     av_pix_fmt_get_chroma_sub_sample(h->avctx->pix_fmt,
@@ -3196,17 +3228,12 @@ static int decode_slice_header(H264Context *h, H264Context *h0)
 
     h->chroma_y_shift = h->sps.chroma_format_idc <= 1; // 400 uses yuv420p
 
-    h->width = 16 * h->mb_width - (2 >> CHROMA444(h)) * FFMIN(h->sps.crop_right, (8 << CHROMA444(h)) - 1);
-    if (h->sps.frame_mbs_only_flag)
-        h->height = 16 * h->mb_height - (1 << h->chroma_y_shift) * FFMIN(h->sps.crop_bottom, (16 >> h->chroma_y_shift) - 1);
-    else
-        h->height = 16 * h->mb_height - (2 << h->chroma_y_shift) * FFMIN(h->sps.crop_bottom, (16 >> h->chroma_y_shift) - 1);
+    h->width  = 16 * h->mb_width;
+    h->height = 16 * h->mb_height;
 
-    if (FFALIGN(h->avctx->width,  16) == h->width &&
-        FFALIGN(h->avctx->height, 16) == h->height) {
-        h->width  = h->avctx->width;
-        h->height = h->avctx->height;
-    }
+    ret = init_dimensions(h);
+    if (ret < 0)
+        return ret;
 
     if (h->sps.video_signal_type_present_flag) {
         h->avctx->color_range = h->sps.full_range ? AVCOL_RANGE_JPEG
@@ -3221,8 +3248,8 @@ static int decode_slice_header(H264Context *h, H264Context *h0)
     }
 
     if (h->context_initialized &&
-        (h->width  != h->avctx->width   ||
-         h->height != h->avctx->height  ||
+        (h->width  != h->avctx->coded_width   ||
+         h->height != h->avctx->coded_height  ||
          needs_reinit)) {
 
         if (h != h0) {
@@ -4611,6 +4638,26 @@ static int get_consumed_bytes(int pos, int buf_size)
     return pos;
 }
 
+static int output_frame(H264Context *h, AVFrame *dst, AVFrame *src)
+{
+    int i;
+    int ret = av_frame_ref(dst, src);
+    if (ret < 0)
+        return ret;
+
+    if (!h->sps.crop)
+        return 0;
+
+    for (i = 0; i < 3; i++) {
+        int hshift = (i > 0) ? h->chroma_x_shift : 0;
+        int vshift = (i > 0) ? h->chroma_y_shift : 0;
+        int off    = ((h->sps.crop_left >> hshift) << h->pixel_shift) +
+            (h->sps.crop_top  >> vshift) * dst->linesize[i];
+        dst->data[i] += off;
+    }
+    return 0;
+}
+
 static int decode_frame(AVCodecContext *avctx, void *data,
                         int *got_frame, AVPacket *avpkt)
 {
@@ -4648,7 +4695,8 @@ out:
             h->delayed_pic[i] = h->delayed_pic[i + 1];
 
         if (out) {
-            if ((ret = av_frame_ref(pict, &out->f)) < 0)
+            ret = output_frame(h, pict, &out->f);
+            if (ret < 0)
                 return ret;
             *got_frame = 1;
         }
@@ -4683,7 +4731,8 @@ out:
             /* Wait for second field. */
             *got_frame = 0;
         } else {
-            if ((ret = av_frame_ref(pict, &h->next_output_pic->f)) < 0)
+            ret = output_frame(h, pict, &h->next_output_pic->f);
+            if (ret < 0)
                 return ret;
             *got_frame = 1;
         }
diff --git a/libavcodec/h264.h b/libavcodec/h264.h
index f5246febc8..484c9d3cfc 100644
--- a/libavcodec/h264.h
+++ b/libavcodec/h264.h
@@ -164,6 +164,8 @@ typedef struct SPS {
     int mb_aff;                        ///< mb_adaptive_frame_field_flag
     int direct_8x8_inference_flag;
     int crop;                          ///< frame_cropping_flag
+
+    /* those 4 are already in luma samples */
     unsigned int crop_left;            ///< frame_cropping_rect_left_offset
     unsigned int crop_right;           ///< frame_cropping_rect_right_offset
     unsigned int crop_top;             ///< frame_cropping_rect_top_offset
@@ -272,6 +274,7 @@ typedef struct H264Context {
 
     int qp_thresh;      ///< QP threshold to skip loopfilter
 
+    /* coded dimensions -- 16 * mb w/h */
     int width, height;
     int linesize, uvlinesize;
     int chroma_x_shift, chroma_y_shift;
diff --git a/libavcodec/h264_ps.c b/libavcodec/h264_ps.c
index 6c895a46ea..c6587c791b 100644
--- a/libavcodec/h264_ps.c
+++ b/libavcodec/h264_ps.c
@@ -413,37 +413,45 @@ int ff_h264_decode_seq_parameter_set(H264Context *h){
 #endif
     sps->crop= get_bits1(&h->gb);
     if(sps->crop){
-        int crop_vertical_limit   = sps->chroma_format_idc  & 2 ? 16 : 8;
-        int crop_horizontal_limit = sps->chroma_format_idc == 3 ? 16 : 8;
-        sps->crop_left  = get_ue_golomb(&h->gb);
-        sps->crop_right = get_ue_golomb(&h->gb);
-        sps->crop_top   = get_ue_golomb(&h->gb);
-        sps->crop_bottom= get_ue_golomb(&h->gb);
+        int crop_left   = get_ue_golomb(&h->gb);
+        int crop_right  = get_ue_golomb(&h->gb);
+        int crop_top    = get_ue_golomb(&h->gb);
+        int crop_bottom = get_ue_golomb(&h->gb);
+
         if (h->avctx->flags2 & CODEC_FLAG2_IGNORE_CROP) {
-            av_log(h->avctx, AV_LOG_DEBUG,
-                   "discarding sps cropping, "
-                   "original values are l:%u r:%u t:%u b:%u\n",
-                   sps->crop_left,
-                   sps->crop_right,
-                   sps->crop_top,
-                   sps->crop_bottom);
+            av_log(h->avctx, AV_LOG_DEBUG, "discarding sps cropping, original "
+                   "values are l:%u r:%u t:%u b:%u\n", crop_left, crop_right,
+                   crop_top, crop_bottom);
 
             sps->crop_left   =
             sps->crop_right  =
             sps->crop_top    =
             sps->crop_bottom = 0;
-        }
-        if(sps->crop_left || sps->crop_top){
-            av_log(h->avctx, AV_LOG_ERROR, "insane cropping not completely supported, this could look slightly wrong ...\n");
-        }
-        if(sps->crop_right >= crop_horizontal_limit || sps->crop_bottom >= crop_vertical_limit){
-            av_log(h->avctx, AV_LOG_ERROR, "brainfart cropping not supported, this could look slightly wrong ...\n");
+        } else {
+            int vsub = (sps->chroma_format_idc == 1) ? 1 : 0;
+            int hsub = (sps->chroma_format_idc == 1 || sps->chroma_format_idc == 2) ? 1 : 0;
+            int step_x = 1 << hsub;
+            int step_y = (2 - sps->frame_mbs_only_flag) << vsub;
+
+            if (crop_left & (0x1F >> (sps->bit_depth_luma > 8)) &&
+                !(h->avctx->flags & CODEC_FLAG_UNALIGNED)) {
+                crop_left &= ~(0x1F >> (sps->bit_depth_luma > 8));
+                av_log(h->avctx, AV_LOG_WARNING, "Reducing left cropping to %d "
+                       "chroma samples to preserve alignment.\n",
+                       crop_left);
+            }
+
+            sps->crop_left   = crop_left   * step_x;
+            sps->crop_right  = crop_right  * step_x;
+            sps->crop_top    = crop_top    * step_y;
+            sps->crop_bottom = crop_bottom * step_y;
         }
     }else{
         sps->crop_left  =
         sps->crop_right =
         sps->crop_top   =
         sps->crop_bottom= 0;
+        sps->crop = 0;
     }
 
     sps->vui_parameters_present_flag= get_bits1(&h->gb);
diff --git a/libavcodec/options_table.h b/libavcodec/options_table.h
index e1b124b31c..5595c0e4f5 100644
--- a/libavcodec/options_table.h
+++ b/libavcodec/options_table.h
@@ -46,6 +46,7 @@ static const AVOption options[]={
        "to minimum/maximum bitrate. Lowering tolerance too much has an adverse effect on quality.",
        OFFSET(bit_rate_tolerance), AV_OPT_TYPE_INT, {.i64 = AV_CODEC_DEFAULT_BITRATE*20 }, 1, INT_MAX, V|E},
 {"flags", NULL, OFFSET(flags), AV_OPT_TYPE_FLAGS, {.i64 = DEFAULT }, 0, UINT_MAX, V|A|E|D, "flags"},
+{"unaligned", "allow decoders to produce unaligned output", 0, AV_OPT_TYPE_CONST, { .i64 = CODEC_FLAG_UNALIGNED }, INT_MIN, INT_MAX, V | D, "flags" },
 {"mv4", "use four motion vectors per macroblock (MPEG-4)", 0, AV_OPT_TYPE_CONST, {.i64 = CODEC_FLAG_4MV }, INT_MIN, INT_MAX, V|E, "flags"},
 {"qpel", "use 1/4-pel motion compensation", 0, AV_OPT_TYPE_CONST, {.i64 = CODEC_FLAG_QPEL }, INT_MIN, INT_MAX, V|E, "flags"},
 {"loop", "use loop filter", 0, AV_OPT_TYPE_CONST, {.i64 = CODEC_FLAG_LOOP_FILTER }, INT_MIN, INT_MAX, V|E, "flags"},
diff --git a/libavcodec/version.h b/libavcodec/version.h
index 394bf1a386..cfcacb7f84 100644
--- a/libavcodec/version.h
+++ b/libavcodec/version.h
@@ -27,7 +27,7 @@
  */
 
 #define LIBAVCODEC_VERSION_MAJOR 55
-#define LIBAVCODEC_VERSION_MINOR  1
+#define LIBAVCODEC_VERSION_MINOR  2
 #define LIBAVCODEC_VERSION_MICRO  0
 
 #define LIBAVCODEC_VERSION_INT  AV_VERSION_INT(LIBAVCODEC_VERSION_MAJOR, \