Tagged: video Toggle Comment Threads | Keyboard Shortcuts

  • s 4:46 PM on 130315 Permalink | Reply
    Tags: , inspiration, , video   

    3D Map Video 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

    Wide Web World from Paul Wex on Vimeo.

    On the process:

    Hi, first I tested the screen recorders – there was a huge difference between them. The iShowU HD had really the best CPU efficiency, so I began to test the highest screen resolution for the recording without too much loss of the fps. That was 1280×720. I tested my browsers too, but there was no significant difference for me, so I stayed at Chrome. After switching in fullscreen mode I selected the recording range on the screen, so there were no panels or symbols on the screen anymore. Unfortunately the crop had to be quite hard.Start-Move&Record-Stop with short keys, meanwhile 100x computer reboot/log off (the browser became slower and slower after minutes!). I recorded as ProRes422HQ at 720p/30fps and converted to 25p via CinemaTools afterwards. Postproduction, edit and colorgrading in FCP7 with standard plugs and Knoll Light EZ Lensflare.

    Then the music production followed…

    Cheers :)

  • s 10:09 PM on 130221 Permalink | Reply
    Tags: , video   

    Joy of Stats 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

  • s 4:29 PM on 130221 Permalink | Reply
    Tags: video   

    FREE Online Stock Video Sites 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

    20 FREE Online Stock Video Sites! | Premiumbeat.com.

  • s 10:35 AM on 130218 Permalink | Reply
    Tags: , , video   

    Universal Subtitles Tool 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77


    Universal Subtitles is an online subtitling/captioning tool for web videos. My colleague Bryan Nunez commented on the potential uses of the service in this post. You can use Universal Subtitles to translate videos that already exist online, e.g, on YouTube or Blip.tv, regardless whether they are in your account or someone else’s account. You can watch a step-by-step video tutorial here that demonstrates how to use the subtitling interface.

    Once the translation is finished, you can (1) share the video with the subtitles by embedding it in a web site, blog post, or a social networking site. Alternatively, (2) you can download the subtitle track to use locally on your computer. The subtitle track is available to download in several formats: SRT, SSA, SBV, DFXP, etc.

    In this post, I will go over the second of the above two options – the various ways you can make use of the downloaded subtitle track. [flickr id=”6871166605″ thumbnail=”small” overlay=”true” size=”medium” group=”” align=”right”]

    I. We most recently used Universal Subtitles to create Arabic subtitles for a training video. I will add a link to it here once the video gets published. Once the subtitles were done by a remote volunteer translator, I downloaded an SRT file from Universal Subtitles and uploaded it to the YouTube account where the video lives (learn how). This way, the translation lives together with the original video and it is available to access directly on YouTube via the CC button, or it can be embedded in a web site. The advantage of embedding via YouTube is that, unlike Universal Subtitles, YouTube does not use Javascript to generate the embed code – Javascript is viewed as a security risk by some platforms.

    [flickr id=”6836521551″ thumbnail=”small” overlay=”true” size=”medium” group=”” align=”left”]

    II. Alternatively, you can use the SRT file with VLC player to play the video with the subtitles locally on your computer.

    Both YouTube and VLC automatically format the subtitles – font, font size, line breaks, etc. Notice how different the same subtitle looks in the two screenshots.

    Automatic formatting works pretty well in most cases. However, translating specialized human rights language usually results in more text than the original English. So, it’s a good idea to format the translated text for an optimal viewing experience.

    However, when I tried to open the SRT file for editing on my computer the Arabic script did not show correctly. To fix this problem, I used a tool called TextWrangler to change the font encoding to Unicode (UTF-16). [flickr id=”6871004135″ thumbnail=”small” overlay=”true” size=”large” group=”” align=”right”]

    III. Now, with the Arabic script showing correctly, I was able to import it in a tool such as Submerge to create a version of the video with the subtitles burned in the image. However, Submerge has limited text formatting options.

    IV. The best way to format the text is to bring it into FinalCut, Premiere, or some specialized subtitling software. However, bringing subtitles into FinalCut Pro is not as straightforward as importing an SRT file to Submerge. To import the subtitle track in FinalCut, you need a specific type of XML file called XML Interchange Format.

    [flickr id=”6836189381″ thumbnail=”small” overlay=”true” size=”medium” group=”” align=”left”] I used a tool called Title Exchange to convert the SRT file (the file I saved with Unicode (UTF-16) font encoding) to XML for FCP. (Remember, I only had to change the font encoding because my subtitles were in Arabic – Latin and Cyrillic scripts should work straight forward). Read instructions on how to create XML for FCP with Title Exchange.

    Then, I imported the XML into FinalCut Pro to edit and format the subtitles for optimal viewing. [flickr id=”6836777553″ thumbnail=”small” overlay=”true” size=”medium” group=”” align=”right”]

    From FinalCut, you can either export a QuickTime with the subtitles burned into the video image or export the subtitle track as an XML file, then use Title Exchange to convert it to STL, a format that is not currently offered by Universal Subtitles, and import it into DVDStudio Pro for DVD authoring. Watch this video tutorial.

    V. Alternatively, you can import the XML file in Adobe Premiere. Also, you can use Title Exchange to convert the XML file to Adobe Encore text file for DVD/Blu-ray authoring.

    Universal Subtitles is a very useful tool for crowd-sourcing translation and subtitling of online videos and we would love to see some additional features that would simplify the above described process – options for text formatting in the subtitling interface – line breaks, font, font size, etc., also, some additional subtitle formats: STL, XML for FCP/Premiere, and Encore text file.

  • s 10:33 AM on 130218 Permalink | Reply
    Tags: , , video   

    Crowd Sourced Subtitling 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77


  • s 7:29 PM on 130217 Permalink | Reply
    Tags: , , video   

    Warp Stabilizer Settings For Adobe Premiere and After Effects 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

    The Best Warp Stabilizer Settings For Adobe Premiere and After Effects CS5.5 and CS6 | Who Is Matt? Matt Johnson Productions.

    Most shaky footage comes from a lack of control of the camera and is especially prevalent in DSLRs due to their small size.  The form factor and weight don’t lend themselves to a steady shot which is why you often see DSLRs decked out with full shoulder rigs, weights, and setups that resemble something more appropriate for fishing than stabilization.  With the popularity of DSLRs as well as the smaller sizes they are making cinema cameras these days, it makes sense that they would need help in software if you want to have any hope of shooting handheld.

    One of my favorite new effects included in Adobe After Effects CS 5.5, CS6, and Adobe Premiere Pro CS6 is the Warp Stabilizer.  For many years if you had shaky footage your only option was to either spend hours tracking a moving object in After Effects with keyframes in the hopes of it working properly, or learning how to use an expensive plugin software such as Mocha.  Thankfully, Adobe realized there was a need for a much quicker solution that would work for the majority of the clips being filmed in the world today.  Thus, Warp Stabilizer was born and now stabilized footage is only a quick click away.

    The good news is that Warp Stabilizer’s default settings work for the majority of the clips that you throw at it.  If you have a decent clip it will usually make it great, and if you have a great clip with maybe a slight bump in it, (think slider with an uneven bit of railing) it will make it look perfect.  All the computing, tracking, and general analyzing happens in the software and you end up with a great clip in around 1 minute.  But, this post isn’t for those people that just throw it on and have it work.  This post is for two kinds of people, those that like to tinker and want the best possible looking clips with ultimate smoothness and stabilization, or those that have thrown the Warp Stabilizer Effect on a clip and had it result in Jello, distortion, and rolling shutter.

    I found that I often deal with Jello and distortion on my clips when I am shooting handheld and if my camera rotates even a slight bit.  I believe this Jello distortion is due to the way the Effect software analyzes the clip.  Simple explanation: Warp Stabilizer actually analyzes the entire clip in 3D space and is able to tell which objects are in the foreground and background.  Sometimes it has issues with differentiating which is where I believe the Jello distortion becomes evident.

    The following are the settings I would recommend trying out if you are wanting to remove the Jello distortion effect from your clip.  These settings are applicable to both Warp Stabilizer in After Effects CS5.5 and CS6, as well as Premiere Pro CS6.

    To add Warp Stabilizer to your clip in After Effects CS5.5 and CS6, select your layer that you want to apply it to, and go to “Effect > Distort > Warp Stabilizer.” In Premiere Pro CS6, select your clip you want to apply it to and go to your“Effects” window and select “Video Effects > Distort > Warp Stabilizer” or search for it in the Effects search box.

    I will go through them in the order that I usually try when I run into problems with Warp Stabilizer.  Each clip you film will be unique and there is no guarantee Warp Stabilizer will treat each one the same.  While it is always better to film your footage as stable as possible, having the ability to fix your clips that have problems is wonderful.

    Settings to Fix Jell0 and Distortion in Warp Stabilizer

    By default, Warp Stabilizer chooses “Smooth Motion” – 50%, with a method of “SubSpace Warp.”  Your video borders framing will show “Stabilize, Crop, Auto-scale”.  If you’re running into problems of your clip becoming distorted and looking like Jello, I would recommend the following methods to try and clean it up.  If one does not work, try the next until you are satisfied with your clip.

    1. Click the “Advanced” arrow and check the “Detailed Analysis” box.  This will require you to re-analyze your footage and it will take longer than before, but sometimes this is a quick fix.

    Detailed Analysis

    2.  Click the “Advanced” arrow and adjust the “Crop Less <-> Smooth More” percentage from “50% down to 5%”in increments of 10.  For example, change it to 40%, let it stabilize, then check the footage if its improved.  Each time you lower this percentage the video will become a bit more shaky, but it will introduce less jello into the image.  It is helpful to turn this down if you have just a slight amount of shaky-ness (perhaps due to handholding your camera), and you want to smooth it out.

    Crop Less Smooth More

    3.  Click the “Stabilization” arrow and change “Smoothness” from 50% to 5%.  This reduces the smoothing that Warp Stabilizer will attempt to apply to your clip, and while it will result in a slightly more shaky shot, I find that my shots often don’t need that much smoothing.  This often cuts out a lot of the distortion and Jello effect.  


    4.  Click the “Advanced” arrow and change “Rolling Shutter Ripple” from “Automatic Reduction” to “Enhanced Reduction.”  This setting is usually only helpful when dealing with Rolling Shutter introduced from the CMOS sensors used in DSLRS and other popular digital cameras, but it doesn’t hurt to try using it.

    Rolling Shutter Ripple

    5. Click the “Stabilization” arrow and change “Method” to “Position.”  Then click “Borders” and change “Framing”to “Stabilize, Synthesize Edges.”  With this setting, Warp Stabilizer will actually create new edges for your footage from existing pixels.  As long as it isn’t being forced to make up too much information it usually does this very well.  The tradeoff is that I almost always have to render my clips before viewing them because of the enhanced processor power required.  Try using Synthesize Edges while changing the Smoothness percentage.  This usually fixes the Jello/distortion in my shots when nothing else will.

    Synthesize Edges

    Hopefully by changing these Warp Stabilizer settings your video will look distortion and Jello free.  Keep in mind though that each video clip is different and will require a different technique.  There are plenty of other settings to tweak when using Warp Stabilizer, be sure to experiment with them as well.  Let me know in the comments below if you find another great Warp Stabilizer technique.

  • s 10:35 PM on 130215 Permalink | Reply
    Tags: computer, video   

    greenscreen keying technique 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

    you may be seeing keylight’s despill efforts, where it tries to remove any green screen color that spills on the subject from the screen. this often looks a bit like noise that gets added to the subject.

    to double check that that is the issue, toggle the ‘view’ from final result to intermediate and see if the problem persists (at the possible expense of green spill on the subject).
    if you still want to suppress spill but try and lessen then noise that the spill suppression is causing, often setting the replacement method (in screen matte settings) from soft color to hard color can help.
    it essentially uses 2 instances of keylight, the first to create a core matte (it calls inside mask), the second to produce the key.

    another good technique is aharon rabinowitz’s ‘super-tight junk mattes’ technique that you can find in the ae tutorials here at the cow.
    and you can use both techniques on the same key to really define the area that will be affected by keylight.
    for more complex spill suppression situations, andrew devis also has a tutorial here.
    you can also find several techniques and tutorials here (and on the web) for creating light wrap when you are compositing your keyed elements into a shot/scene, which will usually really help to make it look natural.
  • s 10:02 PM on 130215 Permalink | Reply
    Tags: , , , reference, video   

    Reference Gear List – Wedding Video Shoot 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

    (3) Canon 5D Mark IIs
    (1) Canon 7D
    (1) GoPro Hero HD with waterproof housing
    GlideCam HD 2000
    5′ Kessler CineSlider with Oracle Motion Controller
    Igus Slider
    DIY Zipline
    10′ Cobra Crane
    7″ Marshall Monitor
    Zacuto Z-Finder

    35mm f/1.4L
    50mm f/1.2L
    85mm f/1.2L
    100mm Macro f/2.8L IS
    135mm f/2L
    16-35mm f/2.8L
    24-105mm f/4L IS
    70-200mm f/2.8L IS

  • s 9:49 PM on 130215 Permalink | Reply
    Tags: 60D, , , , video   

    Canon 60D 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77

    Body probably not as rugged as mag-alloy 50DAF assist only works when flash is up, though flash can be forced offAuto-exposure not reliable in very low lightNo in-camera image stabilization (lens based)18-135mm kit lens has noticeable chromatic aberration and geometric distortion; also doesn’t focus very closelyIn-camera distortion and chromatic aberration correction only available in post-capture RAW processingFlash exposures sometimes inconsistentNo PC Sync connector for external strobesNo continuous autofocus in Movie modeAuto and Incandescent white balance leaves tungsten lighting too warmDefault high ISO noise suppression has trouble with low-contrast areas, especially in the red channelSingle-area full AF lag and burst mode slower than 50DImatest high quality dynamic range score not as good as some competitorsNo AF MicroadjustmentNeed to use Mode dial to switch to Movie modeSluggish buffer clearing times

  • s 4:44 PM on 130215 Permalink | Reply
    Tags: editing, , video   

    About Drop Frame and Non-Drop Frame Timecode 

    Warning: preg_match_all(): Compilation failed: invalid range in character class at offset 7 in /nfs/c01/h08/mnt/31572/domains/book.kmov.org/html/wp-content/themes/p2/inc/mentions.php on line 77


    About Drop Frame and Non-Drop Frame Timecode

    NTSC video (black-and-white) originally had a frame rate of 30 fps, so the timecode counted at 30 fps. However, NTSC color video (the only kind of NTSC video in use today), has a frame rate of 29.97 fps. This subtle difference between 30 fps and 29.97 fps seems practically negligible and, in many cases, ignoring this discrepancy is fine. But not always. What editors needed, especially in expensive broadcast markets, was timecode that accurately reflected the exact duration of a program on tape.

    There are two types of 30 fps timecode for use with NTSC video: non-drop frame and drop frame. Non-drop frame timecode is simple: for every frame of video, there is a corresponding timecode number. The timecode increments without any compensation. In almost all cases, timecode is non-drop frame. In fact, drop frame timecode only matters in the case of NTSC video.

    Drop frame timecode compensates for the fact that the NTSC format has a frame rate of 29.97 fps, which is .03 fps slower than the nearest whole number frame rate of 30 fps. Timecode can only be represented by whole numbers, so timecode numbers are periodically skipped in drop frame timecode. This way, the timecode number always matches the seconds and minutes of video that have played. NTSC can use either drop frame or non-drop frame timecode.

    Important: No video frames are dropped when you use drop frame timecode. Only the associated timecode numbers are skipped.

    You can think of drop frame timecode as being like leap years on the calendar. In the case of leap years, an extra day is added every 4 years except when the year is divisible by 400. This compensates for the fact that the way days are measured and the way years are measured do not align exactly. Even though the difference is slight, an unacceptable error accumulates over time unless regular adjustments are made to the count.

    More About Drop Frame Timecode and the NTSC Frame Rate

    NTSC video has a frame rate of 29.97 fps, but the timecode counts at 30 fps. To better understand this subtle distinction, remember that the main purpose of timecode is to uniquely label and address each video frame, not to tell time (another name for timecode is address code).

    Consider what it would be like if frames were labeled a different way, without any reference to time. For example, if each frame had a unique address coded with five letters of the alphabet, starting at AAAAA, AAAAB, AAAAC, and so on until ZZZZZ, editors would refer to shots and scenes by their individual five-letter codes. A director requesting a particular shot could look in the log notes and tell the editor to find frame ABAAA on a particular tape.

    On tape or disk, each frame lasts 1/29.97 of a second. Since there is an address affixed to each frame, the timecode moves at the same rate as the video (29.97 fps).

    Now, instead of using a five-letter code to uniquely tag each frame, consider using an address code in the format 00:00:00:00. Remember that these numbers don’t reflect time; they are simply unique identifiers. The first frame of NTSC video is labeled 00:00:00:00. The 29th frame is labeled 00:00:00:29, and the 30th frame is labeled 00:00:01:00. Again, just because a frame is labeled 00:00:01:00 does not mean that 1 second has passed. The frame could just as easily have been named AAABD, in which case there would be no temptation to read the label as a time value. Only the frame rate of the video can determine how much time has passed by the 30th frame. In the case of NTSC video, 0.999 seconds have passed by frame 30. By frame 1800, 60.06 seconds have passed.

    Frame count
    Timecode labels (30 fps)
    Time passed (29.97 fps)
    Error between timecode number and real time
    1/30 of a second
    1/29.97 of a second
    = 30/30 of a second
    = 1 second
    = 30/29.97 of a second
    = 1.001 seconds
    0.001 seconds
    = 60/30 of a second
    = 2 seconds
    = 60/29.97 of a second
    = 2.002 seconds
    0.002 seconds
    1800/30 of a second
    = 60 seconds
    = 1 minute
    =1800/29.97 of a second
    = 60.06 seconds
    = 1.001 minutes
    0.001 minutes
    0.06 seconds
    1.8 frames
    = 18,000/30 of a second
    = 600 seconds
    = 10 minutes
    = 18,000/29.97 of second
    = 600.6 seconds
    = 10.01 minutes
    0.01 minutes
    0.6 seconds
    17.9 frames
    = 108,000/30 of a second
    = 3600 seconds
    = 1 hour
    =108,000/29.97 of a second
    = 3603.6 seconds
    = 1.001 hours
    0.001 hours
    3.6 seconds
    107.89 frames

    If you edit an hour-long program on NTSC video, the 30 fps timecode indicates that the last frame of the program is frame 108,000, labeled as timecode 01:00:00:00 (non-drop frame). However, the table above shows that because the video actually runs at 29.97 fps (each frame is slightly longer than if it were running at 30 fps), 1 hour has actually passed at frame 107,892 (3.6 seconds earlier than the 30 fps timecode shows). What editors wanted, particularly in television environments, was a method of frame addressing that accurately reflected how much time had passed.

    Drop frame timecode was invented to compensate for the discrepancy between 29.97 and 30 fps. Every minute except each tenth minute, two timecode numbers are dropped from the timecode count. This drop frame mode of 30 fps timecode remains accurate compared to the actual time passed, with a strange side effect that two numbers each minute vanish from the count.

    The Difference Between Frame Rate and Timecode

    The frame rate of your film or video describes how rapidly frames are photographed or played back. It refers to the physical speed of image capture and playback. Timecode is merely a method of labeling frames with unique identifiers to easily find them again later. It is a convenient way of giving each frame a name that can be referred to later without having to verbally describe and visually search for it. Even though frame rate and timecode are independent, people commonly confuse the two, which can lead to frustrating problems in post-production. Before you start a project, be certain that you understand the difference between these two terms.

Compose new post
Next post/Next comment
Previous post/Previous comment
Show/Hide comments
Go to top
Go to login
Show/Hide help
shift + esc