{
    "info": {
        "author": "Lucas Ou-Yang",
        "author_email": "lucasyangpersonal@gmail.com",
        "bugtrack_url": null,
        "classifiers": [],
        "description": "Pythonic Video summarization\n----------------------------\n\nSick of nlp text summarization? Presenting: **video summarization**!\n\nWe take advantage of youtube's comment timestamp video references to generate\na summarized + shorter video from any youtube video.\n\nA glance:\n---------\n\n.. code-block:: pycon\n\n    >>> import vsummarize\n\n    >>> data = vsummarize.summarize('http://youtube.com/watch?v=...', output='shorter.mp4')\n\nAfter this command, the physical summarized ``.mp4`` is now in the output path \nyou provided. We return some meta data incase you need it.\n\nTimestamps from youtube comments are included and are important b/c our \nalgorithm generates summaries via the comments and their timestamps.\n\n.. code-block:: pycon\n\n    >>> print data.hot_clips\n    [('0:12', '0:16'), ..., ('12:31', '13:01')]\n\n    >>> print data.timestamps \n    ['0:12', '0:12', '0:14', ..., '12:31']\n\n    >>> print data.duration # in seconds\n    65 \n\nA demo:\n-------\n\nI tested this software on a 1 hour long Obama speech.\n\nOriginal video (59 minutes): http://www.youtube.com/watch?v=hed1nP9X7pI\n\nSummarized video (3:30 minutes): http://www.youtube.com/watch?v=aDYDN9lsSHg\n\nA lot of the time, (even in my product www.shorten.tv), \nyou just want a list of hot video clips instead of physically summarizing\na video into a new ``.mp4`` because of the resource consumption.\n\nTo do this, simply ignore the output video file parameter.\n\n.. code-block:: pycon\n\n    >>> data = vsummarize.summarize('http://youtube.com/watch?v=...')\n\n    >>> print data.hot_clips\n    [('0:12', '0:16'), ..., ('12:31', '13:01')]\n\n\nThe physical, summarized ``.mp4`` has NOT been generated. We just\nretrieved a set of meta data of what would have happened if we did\nsummarize it.\n\nTo actually generate a summary, we use **ffmpeg + moviepy**\nalong with the above ``.hot_clips`` video sequences to stitch together the video.\n\nFeatures\n--------\n\n- ``.mp4`` video summarization\n- youtube comments timestamp extraction\n- youtube video hot timestamp extraction\n- youtube video hot sub-clip extraction\n\nGet it now\n----------\n\nBecause I use both OSX and Ubuntu, I have clear instructions on setting\nup this project in both platforms. However, I can't guarantee\nanything for the other platforms besides give installation advice.\n\nWe use ``moviepy``, the python video manipulation library, which in turn depends \non the ``ffmpeg`` library.\n\nBe sure you have `pip <http://www.pip-installer.org/>`_.\n\nThe installation instructions are as follows:\n\n**OSX**:\n\n::\n\n    $ brew install ffmpeg\n\n\n**Ubuntu**:\n\n::\n\n    $ curl -O http://ffmpeg.gusari.org/static/64bit/ffmpeg.static.64bit.2014-02-28.tar.gz\n    $ tar -zxvf ffmpeg.static.64bit.2014-02-28.tar.gz \n    $ sudo mv ffmpeg ffprobe /usr/local/bin/.\n    $ rm ffmpeg.static.64bit.2014-02-28.tar.gz \n\n\nAnd lastly, don't forget to install ``vSummarize`` itself!\n\n::\n\n    $ pip install vsummarize\n\n\n**This app uses the google gdata api**. I have a file named ``settings.py`` which contains\nmy personal api keys. I've removed that file from this repo for obvious\nreasons but i've included a file called ``rename_to_settings.py`` which has two api key\nvalues for you to cleanly fill out. Also, please rename that file to ``settings.py`` after\nyou are finished!\n\n\nWarning\n-------\n\nBecause this is such a resource intensive task & lib (especially if you are\nactually using the summarized ``.mp4`` generation feature), you may notice on a few\nvideos the ``.mp4`` generation fail due to an *OS memory exception*. This means\nthat you don't have the RAM for ``ffmpeg`` to fork processes to subchunk out your video.\n\n\nLicense\n-------\n\nAuthored and maintained by `Lucas Ou-Yang`_.\nShoutout to `Zulko`_ for helping code and giving advice to \nsome parts of this project.\n\nWe use `moviepy`_ and `ffmpeg`_ for video manipulation.\nWe also use google's youtube api.\nPlease feel free to `email & contact me`_ if you run into issues or just would like\nto talk about the future of this library!\n\n.. _`Lucas Ou-Yang`: http://codelucas.com\n.. _`email & contact me`: mailto:lucasyangpersonal@gmail.com\n.. _`moviepy`: https://github.com/Zulko/moviepy \n.. _`ffmpeg`: http://www.ffmpeg.org/ \n.. _`Zulko`: https://github.com/Zulko\n",
        "description_content_type": null,
        "docs_url": null,
        "download_url": "UNKNOWN",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/codelucas/vsummarize",
        "keywords": null,
        "license": "UNKNOWN",
        "maintainer": null,
        "maintainer_email": null,
        "name": "vsummarize",
        "package_url": "https://pypi.org/project/vsummarize/",
        "platform": "UNKNOWN",
        "project_url": "https://pypi.org/project/vsummarize/",
        "project_urls": {
            "Download": "UNKNOWN",
            "Homepage": "https://github.com/codelucas/vsummarize"
        },
        "release_url": "https://pypi.org/project/vsummarize/0.0.1/",
        "requires_dist": null,
        "requires_python": null,
        "summary": "Python video summarization.",
        "version": "0.0.1"
    },
    "last_serial": 2106606,
    "releases": {
        "0.0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "b2f49f8be0cb7829fe5e2c91bb424550",
                    "sha256": "a8f73892bb14a995a0690252d6d733d00e7f29063550aa30c705de5026e391cd"
                },
                "downloads": -1,
                "filename": "vsummarize-0.0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "b2f49f8be0cb7829fe5e2c91bb424550",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 206068,
                "upload_time": "2016-05-09T14:50:00",
                "url": "https://files.pythonhosted.org/packages/3d/fc/800be29747a31ee35fb85c413f8c424de06847c08e86c194e49e7eace890/vsummarize-0.0.1.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "b2f49f8be0cb7829fe5e2c91bb424550",
                "sha256": "a8f73892bb14a995a0690252d6d733d00e7f29063550aa30c705de5026e391cd"
            },
            "downloads": -1,
            "filename": "vsummarize-0.0.1.tar.gz",
            "has_sig": false,
            "md5_digest": "b2f49f8be0cb7829fe5e2c91bb424550",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 206068,
            "upload_time": "2016-05-09T14:50:00",
            "url": "https://files.pythonhosted.org/packages/3d/fc/800be29747a31ee35fb85c413f8c424de06847c08e86c194e49e7eace890/vsummarize-0.0.1.tar.gz"
        }
    ]
}