1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
FFmpeg/tools
Guo, Yejun 50e194e6e1 tools/python: add script to convert TensorFlow model (.pb) to native model (.model)
For example, given TensorFlow model file espcn.pb,
to generate native model file espcn.model, just run:
python convert.py espcn.pb

In current implementation, the native model file is generated for
specific dnn network with hard-code python scripts maintained out of ffmpeg.
For example, srcnn network used by vf_sr is generated with
https://github.com/HighVoltageRocknRoll/sr/blob/master/generate_header_and_model.py#L85

In this patch, the script is designed as a general solution which
converts general TensorFlow model .pb file into .model file. The script
now has some tricky to be compatible with current implemention, will
be refined step by step.

The script is also added into ffmpeg source tree. It is expected there
will be many more patches and community needs the ownership of it.

Another technical direction is to do the conversion in c/c++ code within
ffmpeg source tree. While .pb file is organized with protocol buffers,
it is not easy to do such work with tiny c/c++ code, see more discussion
at http://ffmpeg.org/pipermail/ffmpeg-devel/2019-May/244496.html. So,
choose the python script.

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
2019-07-01 10:23:47 -03:00
..
python tools/python: add script to convert TensorFlow model (.pb) to native model (.model) 2019-07-01 10:23:47 -03:00
.gitignore
aviocat.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
bisect-create Remove the ffserver program 2018-01-06 18:31:37 +00:00
bookmarklets.html
cl2c tools/cl2c: Add a copyright header 2017-11-22 23:55:23 +00:00
clean-diff
coverity.c tools/coverity: Add model for av_realloc 2016-11-30 15:21:57 -08:00
crypto_bench.c tools/crypto_bench: update the comment about build command 2019-05-09 17:03:30 +08:00
cws2fws.c
dvd2concat
enum_options.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
ffescape.c
ffeval.c
ffhash.c
fourcc2pixfmt.c tools/fourcc2pixfmt: fix usages of av_get_codec_tag_string() 2017-03-29 14:49:29 +02:00
gen-rc
graph2dot.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
ismindex.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
libav-merge-next-commit
loudnorm.rb tools: add loudnorm script example to use loudnorm 2016-11-11 19:22:52 +01:00
make_chlayout_test
Makefile tools: Add fuzzer for demuxers 2019-05-31 22:44:30 +02:00
missing_codec_desc
murge
normalize.py
patcheck patcheck: Add 'threshhold' to common typo list 2017-12-13 12:25:44 +01:00
pktdumper.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
plotframes
probetest.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
qt-faststart.c tools/qt-faststart: Allow free atoms after moov atom. 2018-08-01 00:49:54 +02:00
seek_print.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
sidxindex.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
sofa2wavs.c tools/sofa2wavs: add license header 2017-06-12 22:14:30 +02:00
target_dec_fate.list tools/target_dec_fate.list: add issues 4000 to 6000 2019-03-20 21:12:45 +01:00
target_dec_fate.sh tools/target_dec_fate.sh: Add support for lines that are comments 2019-03-01 23:25:08 +01:00
target_dec_fuzzer.c tools/target_dec_fuzzer: Also fuzz extradata 2019-06-14 21:36:39 +02:00
target_dem_fuzzer.c tools: Add fuzzer for demuxers 2019-05-31 22:44:30 +02:00
trasher.c
uncoded_frame.c lav*,tests: remove several register_all calls 2018-04-02 03:26:22 +01:00
unwrap-diff
yuvcmp.c
zmqsend.c tools/zmqsend: Do not truncate fgetc() return 2016-12-24 14:46:25 +01:00
zmqshell.py