From: Niklas Haas <g...@haasn.dev>
The current logic hard-coded a check for v_sub == 1. We can extend this
logic slightly to cover the case of interlaced 4:1:0 (which has v_sub ==
2).
Here is a diagram explaining this scenario (with center-siting):
a a a a a a a a
b b b b b b b b
X X
a a a a a a a a
b b b b b b b b
a a a a a a a a
b b b b b b b b
Y Y
a a a a a a a a
b b b b b b b b
a = even luma rows
b = odd luma rows
X = even chroma sample
Y = odd chroma sample
In progressive mode, the chroma samples sit at (384, 384) respectively.
Relative to the 8x4 grid of even luma samples (a), the X sample sits at:
h_chr_pos = 384
v_chr_pos = 192
Relative to the 8x4 grid of odd luma samples (b), the Y sample sits at:
h_chr_pos = 384
v_chr_pos = 576
The new code calculates the correct values in all circumstances.
---
libavfilter/vf_scale.c | 16 +++++++++++++---
1 file changed, 13 insertions(+), 3 deletions(-)
diff --git a/libavfilter/vf_scale.c b/libavfilter/vf_scale.c
index b2c9d0b187..2d82141b70 100644
--- a/libavfilter/vf_scale.c
+++ b/libavfilter/vf_scale.c
@@ -642,9 +642,19 @@ static void calc_chroma_pos(int *h_pos_out, int
*v_pos_out, int chroma_loc,
v_pos = v_pos_override;
/* Fix vertical chroma position for interlaced frames */
- if (v_sub == 1 && index > 0) {
- v_pos += 256 * (index == 2); /* offset by one luma row for odd rows */
- v_pos >>= 1; /* double luma row distance */
+ if (v_sub && index > 0) {
+ /* When vertically subsampling, chroma samples are effectively only
+ * placed next to even rows. To access them from the odd field, we need
+ * to account for this shift by offsetting the distance of one luma
row.
+ *
+ * For 4x vertical subsampling (v_sub == 2), they are only placed
+ * next to every *other* even row, so we need to shift by three luma
+ * rows to get to the chroma sample. */
+ if (index == 2)
+ v_pos += (256 << v_sub) - 256;
+
+ /* Luma row distance is doubled for fields, so halve offsets */
+ v_pos >>= 1;
}
/* Explicitly strip chroma offsets when not subsampling, because it
--
2.45.2
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
