1 /*! \page usage_decode Decoding
3 The aom_codec_decode() function is at the core of the decode loop. It
4 processes packets of compressed data passed by the application, producing
5 decoded images. The decoder expects packets to comprise exactly one image
6 frame of data. Packets \ref MUST be passed in decode order. If the
7 application wishes to associate some data with the frame, the
8 <code>user_priv</code> member may be set. The <code>deadline</code>
9 parameter controls the amount of time in microseconds the decoder should
10 spend working on the frame. This is typically used to support adaptive
11 \ref usage_postproc based on the amount of free CPU time. For more
12 information on the <code>deadline</code> parameter, see \ref usage_deadline.
17 \section usage_cb Callback Based Decoding
18 There are two methods for the application to access decoded frame data. Some
19 codecs support asynchronous (callback-based) decoding \ref usage_features
20 that allow the application to register a callback to be invoked by the
21 decoder when decoded data becomes available. Decoders are not required to
22 support this feature, however. Like all \ref usage_features, support can be
23 determined by calling aom_codec_get_caps(). Callbacks are available in both
24 frame-based and slice-based variants. Frame based callbacks conform to the
25 signature of #aom_codec_put_frame_cb_fn_t and are invoked once the entire
26 frame has been decoded. Slice based callbacks conform to the signature of
27 #aom_codec_put_slice_cb_fn_t and are invoked after a subsection of the frame
28 is decoded. For example, a slice callback could be issued for each
29 macroblock row. However, the number and size of slices to return is
30 implementation specific. Also, the image data passed in a slice callback is
31 not necessarily in the same memory segment as the data will be when it is
32 assembled into a full frame. For this reason, the application \ref MUST
33 examine the rectangles that describe what data is valid to access and what
34 data has been updated in this call. For all their additional complexity,
35 slice based decoding callbacks provide substantial speed gains to the
36 overall application in some cases, due to improved cache behavior.
39 \section usage_frame_iter Frame Iterator Based Decoding
40 If the codec does not support callback based decoding, or the application
41 chooses not to make use of that feature, decoded frames are made available
42 through the aom_codec_get_frame() iterator. The application initializes the
43 iterator storage (of type #aom_codec_iter_t) to NULL, then calls
44 aom_codec_get_frame repeatedly until it returns NULL, indicating that all
45 images have been returned. This process may result in zero, one, or many
46 frames that are ready for display, depending on the codec.
49 \section usage_postproc Postprocessing
50 Postprocessing is a process that is applied after a frame is decoded to
51 enhance the image's appearance by removing artifacts introduced in the
52 compression process. It is not required to properly decode the frame, and
53 is generally done only when there is enough spare CPU time to execute
54 the required filters. Codecs may support a number of different
55 postprocessing filters, and the available filters may differ from platform
56 to platform. Embedded devices often do not have enough CPU to implement
57 postprocessing in software. The filter selection is generally handled
58 automatically by the codec, depending on the amount of time remaining before
59 hitting the user-specified \ref usage_deadline after decoding the frame.