 Tim Ledbetter
|
569e7173cc
LibVideo/VP9: Avoid integer overflow during in place butterfly rotation
|
hai 1 ano |
|
Timothy Flynn
|
7c4b0b0389
LibGfx+Userland: Rename Size::scaled_by to Size::scaled
|
hai 1 ano |
|
Zaggy1024
|
888af08638
LibVideo/VP9: Implement lossless residual decoding
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
2ec043c4db
LibVideo/VP9: Make inter-prediction fast path accumulators 32-bit
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
d6b867ba89
LibVideo/VP9: Force inlining of `inverse_transform_2d()` and the IDCT
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
90c0e1ad8f
LibVideo/VP9: Pre-calculate the quantizers at the start of each frame
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
094b0d8a78
LibVideo/VP9: Use an enum to select segment features
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
6e6cc1ddb2
LibVideo/VP9: Make a lookup table for bit reversals
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
f6764beead
LibVideo/VP9: Specialize transforms on their block size
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
8ad0dff5c2
LibVideo/VP9: Implement unscaled fast paths in inter prediction
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
8cd72ad1ed
LibVideo/VP9: Use the Y scale value in `predict_inter_block()`
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
f2c0cee522
LibVideo/VP9: Consolidate frame size calculations
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
57c7389200
LibVideo/VP9: Fix rounding of components in the motion vector selection
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
1fcac52e77
LibVideo/VP9: Count syntax elements in TileContext, and sum at the end
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
5e3192c8d9
LibVideo/VP9: Extend the borders on reference frames to avoid branching
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
08b90bb2d0
LibVideo/VP9: Clamp reference frame prediction coords outside loops
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
bc49af08b4
LibVideo/VP9: Pre-calculate inter-frames' reference frame scale factors
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
5cd5edc3bd
LibVideo/VP9: Copy data to reference frames row by row
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
e6c3b0e495
LibVideo/VP9: Rename `round_2()` to `rounded_right_shift()` for clarity
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
33ff3427eb
LibVideo/VP9: Drop the decoder intermediate bounds checks
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
7b92eff4a6
LibVideo/VP9: Use u32 to store the parsed value counts
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
f58c5ff569
LibVideo/VP9: Correct the mode/partition probability adaption counts
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
4224f253af
LibVideo/VP9: Increase the size of summed boolean counts in merge_probs
|
%!s(int64=2) %!d(string=hai) anos |
|
Linus Groh
|
9c08bb9555
AK: Remove `try_` prefix from FixedArray creation functions
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
f5ea6c89df
LibVideo/VP9: Put reference frames into a struct
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
316dad7bf7
LibVideo/VP9: Remove m_tokens and m_token_cache from Parser
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
a4f14f220d
LibVideo/VP9: Fully qualify all reference frame type enum values
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
db9f1a18f8
LibVideo/VP9: Convert TransformMode to an enum class
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
1a2d8ac40c
LibVideo/VP9: Prefix TransformSize with Transform_ instead of TX_
|
%!s(int64=2) %!d(string=hai) anos |
|
Zaggy1024
|
f6e645a153
LibVideo/VP9: Rename TX(Mode|Size) to Transform(Mode|Size)
|
%!s(int64=2) %!d(string=hai) anos |
