1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 20:37:35 +00:00
Commit graph

45 commits

Author SHA1 Message Date
Zaggy1024
fedbc12c4d LibVideo/VP9: Move segmentation parameters to FrameContext
Note that some of the previous segmentation feature settings must be
preserved when a frame is decoded that doesn't use segmentation.

This change also allowed a few functions in Decoder to be made static.
2022-11-30 08:28:30 +01:00
Zaggy1024
f4761dab09 LibVideo/VP9: Index inter-frame references with named fields or an enum
Previously, we were using size_t, often coerced from bool or u8, to
index reference pairs. Now, they must either be taken directly from
named fields or indexed using the `ReferenceIndex` enum with options
`primary` and `secondary`. With a more explicit method of indexing
these, the compiler can aid in using reference pairs correctly, and
fuzzers may be able to detect undefined behavior more easily.
2022-11-30 08:28:30 +01:00
Zaggy1024
3af4deba6d LibVideo/VP9: Move reference frame type fields to FrameContext 2022-11-30 08:28:30 +01:00
Zaggy1024
b966f9d811 LibVideo/VP9: Move the transform mode field from Parser to FrameContext 2022-11-30 08:28:30 +01:00
Zaggy1024
396972bb69 LibVideo/VP9: Retain adjacent block contexts storage between frames
Re-allocating the storage is unnecessary, since the size will rarely
change during playback.
2022-11-30 08:28:30 +01:00
Zaggy1024
ea7a6f343b LibVideo/VP9: Select and read motion vectors without fields in Parser
Candidate vector selections are only used to calculate the new vectors
for the current block, so we only need to keep those for the duration
of the inter_block_mode_info() call.

Candidate vectors are now stored in BlockMotionVectorCandidates, which
contains the fields necessary to choose the vector to use to sample
from the selected reference frame.

Most functions related to motion vectors were renamed to more verbose
but meaningful names.
2022-11-30 08:28:30 +01:00
Zaggy1024
368687a74f LibVideo/VP9: Store tile counts in FrameContext
The log2 of tile counts in the horizontal and vertical dimensions are
now stored in the FrameContext struct to be kept only as long as they
are needed.
2022-11-30 08:28:30 +01:00
Zaggy1024
6533c5f6a8 LibVideo/VP9: Move more block fields into the BlockContext struct
This includes the segment IDs, transform block sizes, prediction modes,
sub-block counts, interpolation filters and sub-block motion vectors.
2022-11-30 08:28:30 +01:00
Zaggy1024
f4e835635f LibVideo/VP9: Move quantizer indices into FrameContext
This also renames (most?) of the related quantizer functions and
variables to make more sense. I haven't determined what AC/DC stands
for here, but it may be just an arbitrary naming scheme for the first
and subsequent coefficients used to quantize the residuals for a block.
2022-11-30 08:28:30 +01:00
Zaggy1024
0df5c1f32f LibVideo/VP9: Move loop filter parameters to FrameContext 2022-11-30 08:28:30 +01:00
Zaggy1024
90f16c78fa LibVideo/VP9: Move fields set in uncompressed_header() to FrameContext 2022-11-30 08:28:30 +01:00
Zaggy1024
40bc987fe3 LibVideo/VP9: Store color config in the frame context
The color config is reused for most inter predicted frames, so we use a
struct ColorConfig to store the config from intra frames, and put it in
a field in Parser to copy from when an inter frame without color config
is encountered.
2022-11-30 08:28:30 +01:00
Zaggy1024
3259c99cab LibVideo/VP9: Choose whether/how to show new frames using an enum
There are three mutually exclusive frame-showing states:
- Show no new frame, only store the frame as a reference.
- Show a newly decoded frame.
- Show frame from the reference frame store.
Since they are mutually exclusive, using an enum rather than two bools
makes more sense.
2022-11-30 08:28:30 +01:00
Zaggy1024
befcd479ae LibVideo/VP9: Add Frame, Tile and Block context structs
These are used to pass context needed for decoding, with mutability
scoped only to the sections that the function receiving the contexts
needs to modify. This allows lifetimes of data to be more explicit
rather than being stored in fields, as well as preventing tile threads
from modifying outside their allowed bounds.
2022-11-30 08:28:30 +01:00
Zaggy1024
9da432f4d6 LibVideo/VP9: Remove m_eob_total field from parser
The field was only used once to track whether residual tokens were
present in the block. Parser::tokens() now returns a bool indicating
whether they were present.
2022-11-30 08:28:30 +01:00
Zaggy1024
10d207959d LibVideo/VP9: Remove m_mi_row and col fields from the parser
These are now passed as parameters to each function that uses them.
These will later be moved to a struct to further reduce the amount of
parameters that get passed around.

Above and left per-frame block contexts are now also parameters passed
to the functions that use them instead of being retrieved when needed
from a field. This will allow them to be more easily moved to a tile-
specific context later.
2022-11-30 08:28:30 +01:00
Zaggy1024
4a4aa697d9 LibVideo/VP9: Use a struct for block context to keep between frames
There are three fields that we need to store from FrameBlockContext to
keep between frames, which are used to parse for those same fields for
the next frame.
2022-11-30 08:28:30 +01:00
Zaggy1024
5275a1101e LibVideo/VP9: Remove dump_frame_info() function from Decoder
The function serves no purpose now, any debug information we want to
pull from the decoder should be instead accessed by some other yet to
be created interface.
2022-11-30 08:28:30 +01:00
Zaggy1024
0638c5d2b8 LibVideo/VP9: Use a class to store 2D context information 2022-11-30 08:28:30 +01:00
Zaggy1024
44413c31a9 LibVideo/VP9: Store data used between decode_block calls in a struct
All state that needed to persist between calls to decode_block was
previously stored in plain Vector fields. This moves them into a struct
which sets a more explicit lifetime on that data. It may be possible to
store this data on the stack of a function with the appropriate
lifetime now that it is split into its own struct.
2022-11-30 08:28:30 +01:00
Zaggy1024
9b6ab1d4e5 LibVideo/VP9: Change fields within decode_partition() to variables 2022-11-30 08:28:30 +01:00
Zaggy1024
e379223633 LibVideo/VP9: Don't store the default_intra_mode in a field
The default intra prediction mode was only used to set the sub-block
modes and the y prediction mode. Instead of storing it in a field, with
the sub modes are stored in an Array, we can just pull the last element
to set the y mode.
2022-11-30 08:28:30 +01:00
Zaggy1024
713b48cfe2 LibVideo/VP9: Remove unused parser field m_is_compound 2022-11-30 08:28:30 +01:00
Zaggy1024
907816e629 LibVideo: Create TokensContext struct for token parsing parameters
With the addition of this struct, both the bool to determine if coefs
should be parsed and the token parse itself can take specific
parameters.

This is the last step in parameterizing all the tree parsing, so the
old functions in TreeParser are now unused. This patch is very
satisfying :^)

There's still more work to be done to clean up how the parameters are
passed from Parser, but that's work for another day.
2022-11-12 10:17:27 -07:00
Zaggy1024
c6ecad63d0 LibVideo: Parameterize parsing of the default VP9 intra mode tree 2022-11-12 10:17:27 -07:00
Zaggy1024
fce7639c90 LibVideo: Put motion vector or reference frame pairs in a struct
Since these two types are often passed around as a pair, it's easier to
handle them with a simple pair struct, at least for now. Once things
are fully being passed around as parameters wherever possible, it may
be good to change this type for something more generalized.
2022-11-12 10:17:27 -07:00
Zaggy1024
17e1b205a4 LibVideo: Use the BlockSubsize enum where appropriate in the VP9 parser 2022-11-12 10:17:27 -07:00
Zaggy1024
981997c039 LibVideo: Combine VP9's Intra- and InterMode enums into PredictionMode
The two different mode sets are stored in single fields, and the
underlying values didn't overlap, so there was no reason to keep them
separate.

The enum is now an enum class as well, to enforce that almost all uses
of the enum are named. The only case where underlying values are used
is in lookup tables, but it may be worth abstracting that as well to
make array bounds more clear.
2022-11-12 10:17:27 -07:00
Zaggy1024
1c6d0a9777 LibVideo: Use Gfx::Size for VP9 frame sizes
Frame sizes will now be represented by Gfx::Size instead of storing
width and height separately.
2022-11-12 10:17:27 -07:00
Zaggy1024
40b0bb0914 LibVideo: Change all Span<u8 const> to ReadonlyBytes 2022-11-12 10:17:27 -07:00
Zaggy1024
993385f18d LibVideo: Rename VP9's ReferenceFrame enum to ReferenceFrameType 2022-11-12 10:17:27 -07:00
Zaggy1024
cd127b65c3 LibVideo: Implement CICP color space conversion
This adds a struct called CodingIndependentCodePoints and related enums
that are used by video codecs to define its color space that frames
must be converted from when displaying a video.

Pre-multiplied matrices and lookup tables are stored to avoid most of
the floating point division and exponentiation in the conversion.
2022-10-25 11:06:11 +02:00
Zaggy1024
41cb705b47 LibVideo: Allow the VP9 decoder to decode ultra high resolution video
Previously, some integer overflows and truncations were causing parsing
errors for 4K videos, with those fixed it can fully decode 8K video.

This adds a test to ensure that 4K video will continue to be decoded.

Note: There seems to be unexpectedly high memory usage while decoding
them, causing 8K video to require more than a gigabyte of RAM. (!!!)
2022-10-12 00:54:31 -06:00
Zaggy1024
be0760871e LibVideo: Add support for VP9 superframes
This allows the second shown frame of the VP9 test video to be decoded,
as the second chunk uses a superframe to encode a reference frame and
a second to inter predict between the keyframe and the reference frame.
2022-10-09 20:32:40 -06:00
Zaggy1024
b0187dfc27 LibVideo: Implement inter prediction
This enables the second frame of the test video to be decoded.

It appears that the test video uses a superframe (group of multiple
frames) for the first chunk of the file, but we haven't implemented
superframe parsing.

We also ignore the show_frame flag, so for now, this
means that the second frame read out is shown when it should not be. To
fix this, another error type needs to be implemented that is "thrown" to
decoder's client so they know to send another sample buffer.
2022-10-09 20:32:40 -06:00
Zaggy1024
03738aa006 LibVideo: Implement block parsing for inter frames
This gets the decoder closer to fully parsing the second frame without
any errors. It will still be unable to output an inter-predicted frame.
The lack of output causes VideoPlayer to crash if it attempts to read
the buffers for frame 1, so it is still limited to the first frame.
2022-10-09 20:32:40 -06:00
Zaggy1024
1dc4652683 LibVideo: Rename MV to MotionVector for clarity 2022-10-09 20:32:40 -06:00
Zaggy1024
1514004cd5 LibVideo: Implement VP9 intra-predicted frame decoding
The first keyframe of the test video can be decoded with these changes.

Raw memory allocations in the Parser have been replaced with Vector or
Array to avoid memory leaks and OOBs.
2022-10-09 20:32:40 -06:00
Zaggy1024
da9ff31166 LibVideo: Make new DecoderError class to report useful errors
This allows runtime strings, so we can format the errors to make them
more helpful. Errors in the VP9 decoder will now print out a function,
filename and line number for where a read or bitstream requirement
has failed.

The DecoderErrorCategory enum will classify the errors so library users
can show general user-friendly error messages, while providing the
debug information separately.

Any non-DecoderErrorOr<> results can be wrapped by DECODER_TRY to
return from decoder functions. This will also add the extra information
mentioned above to the error message.
2022-10-09 20:32:40 -06:00
Zaggy1024
b37ea6b414 LibVideo: Allow bit stream reads to throw errors
Errors are propagated to the user of the decoder so that they can be
aware of specific places where a read failed.
2022-10-09 20:32:40 -06:00
FalseHonesty
d60bd42972 LibVideo/VP9: Implement MV reading & rectify MV storage issues
With this patch we are finally done with section 6.4.X of the spec :^)
The only parsing left to be done is 6.5.X, motion vector prediction.

Additionally, this patch fixes how MVs were being stored in the parser.
Originally, due to the spec naming two very different values very
similarly, these properties had totally wrong data types, but this has
now been rectified.
2021-07-10 21:28:56 +02:00
FalseHonesty
27fdf8361c LibVideo/VP9: Finish implementing block decoding (6.4.4)
Though technically block decoding calls into some other incomplete
methods, so it isn't functionally complete yet. However, we are
very close to being done with the 6.4.X sections :)
2021-07-10 21:28:56 +02:00
FalseHonesty
aa27ca1b16 LibVideo/VP9: Implement token parsing (6.4.24-6.4.26)
Note that this now requires a couple new syntax types to be parsed
in the TreeParser, so a follow-up commit will implement that behavior.
2021-07-10 21:28:56 +02:00
FalseHonesty
d79c9c262f LibVideo/VP9: Implement sections 6.1.2 and 8.4.1-8.4.4
These section implement the behavior to refresh the probability
tables after parsing a frame.
2021-07-10 21:28:56 +02:00
FalseHonesty
514559f074 LibVideo/VP9: Rename Decoder -> Parser & create an actual Decoder class
The class that was previously named Decoder handled section 6.X.X of
the spec, which actually deals with parsing out the syntax of the data,
not the actual decoding logic which is specified in section 8.X.X.
The new Decoder class will be in charge of owning and running the
Parser, as well as implementing all of the decoding behavior.
2021-07-10 21:28:56 +02:00