graphics/decord: New port: Efficient video loader for deep learning

This commit is contained in:
Yuri Victorovich 2024-07-16 21:28:22 -07:00
parent 710d523999
commit bf413b3a6b
5 changed files with 68 additions and 0 deletions

View file

@ -34,6 +34,7 @@
SUBDIR += cmrtlib
SUBDIR += cx88
SUBDIR += dav1d
SUBDIR += decord
SUBDIR += deforaos-player
SUBDIR += dirac
SUBDIR += dragon

View file

@ -0,0 +1,31 @@
PORTNAME= decord
DISTVERSIONPREFIX= v
DISTVERSION= 0.6.0-7
DISTVERSIONSUFFIX= -gd2e5619
CATEGORIES= multimedia # machine-learning
PATCH_SITES= https://github.com/${GH_ACCOUNT}/${GH_PROJECT}/commit/
PATCHFILES= e543d1cbd4d21dbac85b1e78e43f59495c8ce78b.patch:-p1 # https://github.com/dmlc/decord/pull/309
MAINTAINER= yuri@FreeBSD.org
COMMENT= Efficient video loader for deep learning
WWW= https://github.com/dmlc/decord
LICENSE= APACHE20
LICENSE_FILE= ${WRKSRC}/LICENSE
LIB_DEPENDS= libavcodec.so:multimedia/ffmpeg \
libdmlc.so:devel/dmlc-core
USES= cmake
USE_GITHUB= yes
GH_ACCOUNT= dmlc
GH_TUPLE= dmlc:dlpack:5c792ce:dlpack/3rdparty/dlpack
post-install:
# install headers
@cd ${WRKSRC} && \
${COPYTREE_SHARE} include ${STAGEDIR}${PREFIX}
.include <bsd.port.mk>

View file

@ -0,0 +1,7 @@
TIMESTAMP = 1721190094
SHA256 (dmlc-decord-v0.6.0-7-gd2e5619_GH0.tar.gz) = a69c60a41075e5044f7d20484aaf04f1437948af3d1df136f6f8b8b7b8c02404
SIZE (dmlc-decord-v0.6.0-7-gd2e5619_GH0.tar.gz) = 19223029
SHA256 (dmlc-dlpack-5c792ce_GH0.tar.gz) = dbcbe1d5ddf8596e18d62906e5467ee2243bb3aff41623fac8dc1ef350cd7539
SIZE (dmlc-dlpack-5c792ce_GH0.tar.gz) = 37929
SHA256 (e543d1cbd4d21dbac85b1e78e43f59495c8ce78b.patch) = 924b59877d0f2890f216b8c5dce52e4fd165ae7fe6a3207231178a18e05fca48
SIZE (e543d1cbd4d21dbac85b1e78e43f59495c8ce78b.patch) = 3846

View file

@ -0,0 +1,13 @@
Decord is a reverse procedure of Record. It provides convenient video slicing
methods based on a thin wrapper on top of hardware accelerated video decoders,
e.g.
FFMPEG/LibAV(Done)
Nvidia Codecs(Done)
Intel Codecs
Decord was designed to handle awkward video shuffling experience in order to
provide smooth experiences similar to random image loader for deep learning.
Decord is also able to decode audio from both video and audio files. One can
slice video and audio together to get a synchronized result; hence providing
a one-stop solution for both video and audio decoding.

View file

@ -0,0 +1,16 @@
include/decord/audio_interface.h
include/decord/av_interface.h
include/decord/base.h
include/decord/runtime/c_backend_api.h
include/decord/runtime/c_runtime_api.h
include/decord/runtime/device_api.h
include/decord/runtime/module.h
include/decord/runtime/ndarray.h
include/decord/runtime/node_base.h
include/decord/runtime/packed_func.h
include/decord/runtime/registry.h
include/decord/runtime/serializer.h
include/decord/runtime/threading_backend.h
include/decord/runtime/util.h
include/decord/video_interface.h
lib/libdecord.so