Video and media guide

This document has been prepared so that authors and researchers can optimize the video content they supply OUP for use within digitally delivered articles. The typical scenario would involve an author with some video knowledge who has a need to film, edit, and encode video material to illustrate their written text. This content could consist of interviews, experiments, technical and medical filming as well as graphics and PowerPoint slides sequenced with audio. We have worked with a great deal of user-generated video over the past few years and the aim of this document is to provide tips on how to avoid common pitfalls in shooting and editing. We also provide some information on how best to encode for OUP’s video delivery platforms.

Equipment choice

There are a wide variety of relatively inexpensive video cameras that are capable of giving very good results. We would recommend using a high-definition camera with a 16:9 widescreen aspect ratio, ideally recording to a memory card. This will not only produce good quality future-proofed footage, but also interface well with OUP’s current video delivery systems. You can record in standard definition on a DV tape-based camera but this type of camera is getting less popular as time goes on. Ideally, you should always mount the camera on a tripod when you film. Shaky hand-held footage is not only hard to watch but won’t encode easily due to the large amount of random movement.

Some attention should also be paid to audio considerations. Many consumer cameras do not have additional audio inputs that enable the user to plug in an external microphone. As a result you are forced to rely on the internal camera microphone which can be inherently noisy and will also pick up more room reverberation and extraneous sounds from outside the room. If you are filming interviews or voice-over, an ideal solution is to use a lapel or tiepin microphone that you can plug directly into the camera, enabling you to capture good-quality close sound. Consumer cameras tend to have a mini-jack socket whereas more professional cameras will have XLR sockets. Do be sure that the microphone you source has the correct connector!

When filming you can either use ‘available light’ or introduce your own additional lighting into the location. In some scenarios, the restrictions of the location will mean that you simply can’t add any additional lighting. In this scenario it is a good idea to switch on as many lights as you can within the room. Modern video cameras will work well in low light but tend to introduce gain amplification into the image if the lighting is too low. This adds video noise into the picture and can make it more difficult to edit and encode due to the lack of contrast. If you are thinking about adding additional lighting, the best results will be achieved through adding broad soft lighting rather than harsh spotlights.

Choice of filming location

In general, you should aim to film in a quiet location with simple backgrounds. When choosing a location, listen carefully to what sounds you can hear in the area – for instance, air-conditioning, aircraft and traffic noises. Also check that the noise isn’t intermittent – for instance a refrigerator switching on and off. This can make it difficult to edit smoothly between sections as the sound will be present in some parts of the video and absent in others. If you have to film a demonstration in a noisy environment then it can be a good idea to record the voice-over explanation in a quieter place afterwards and edit the two together in post-production.

You should also think carefully about what is behind the subject you are filming. Be very cautious filming anything with a window behind as you run the risk of the subject becoming silhouetted on the video. The reason for this is that the light coming from outside is of a much higher intensity than the light inside. Video cameras will often adjust their exposure to accommodate the stronger outside light, but this will result in the under-exposure of the subject inside. You can get around this by shutting curtains and blinds or by adding more light within the room so it balances better with the light outside.

Filming anything in daylight – either outside or in rooms with a lot of glass – can also be problematic as light levels tend to rise and fall in intensity as clouds move across the sun. This can make it very difficult to edit between different parts of the video as in some the light level is going to be quite high and in others it will be relatively low. If filming with daylight you should also avoid shooting with the sunlight behind the subject or you run the risk of them appearing in shadow. It is better to move the camera round so that the sun is lighting the person from the front and to the side.

Framing and shot size

One of the biggest problems we see with self-generated footage is poor framing of the subject within the video screen. If you are filming an interview then make sure that there isn’t unnecessary headroom above the interviewee’s head. Ideally you should imagine the screen split horizontally into three equal portions with the eyes appearing on the line between the top and middle thirds. You should also vary the shot size to add interest throughout the interview – for instance you might want to film the first question on a wide shot and the second question on a closer shot. Alternatively you could run through the interview once on a wide shot and then repeat all of the questions for a second time on a close shot. This will give you a great deal of flexibility when editing.

You should also give some thought as to where the eye-line of the subject goes. Videos can be presented straight to camera with the subject looking directly into the camera lens. If you are filming an interview rather than a presentation, then it is common for the questioner to sit out of shot, to the side and slightly behind the camera. It makes for a more natural feel if the interviewee then looks directly at the questioner rather than into the camera. If you are filming this way, do make sure that the questioner’s notes or the side of their face don’t periodically appear on the edge of the shot.

Likewise, when filming experiments or other technical procedures, you should think carefully about varying the shot sizes and framing for variety and interest. In general it is a good idea to film a wide establishing shot that will enable the viewer to see the totality of the action. Make sure the subject in this wide shot fills the frame with as little space around as possible. Once you have an establishing shot, you can then zoom in to capture more of the detailed action. Try not to introduce too much unnecessary zooming as this can be distracting for the viewer and makes for less efficient encoding. In general it is better to cut between the two shot sizes when you edit rather than letting the viewer see the movement in and out.

If you are filming a subject close up and find that the camera is out of focus, try moving the camera further away from the subject and then zooming in. Finally, if you are filming medical procedures, don’t forget that certain types of equipment also have a video or stills capture facility which can add a great amount of detail to your edit.

Interview technique

If you are filming interviews, you have a choice to make about whether you include the interviewer in your video. In a traditional interview scenario the illusion is constructed that the subject is responding to the questions of an unseen interviewer. Apart from the interviewer being out of shot behind the camera, their questions never actually appear in the final video, rather the interviewee is encouraged to frame the question within their answer.

For instance, if the interviewer asked the question, ‘What is the best thing about working in Oxford?’ then the correct response from the subject would be, ‘The best thing about working in Oxford is…’ or ‘One thing I like about working in Oxford is…’ As a result, when the questions are cut out in the editing process there is context to the answer being given. If you are using this approach then be careful that the interviewer doesn’t audibly agree with the subject – for instance by saying ‘yes’ or ‘mmm’. They can certainly be visually enthusiastic – for instance nodding and smiling – but any audio intrusion ruins the illusion of the unseen interviewer.

An alternative approach is for the interviewer to appear in the video. This can work well if you are using two cameras – for instance a wide shot on the interviewer and interviewee, and a second close-up shot on the interviewee. You can also do this with one camera if you have time to run through the questions once on a wide shot and then once on a close-up. If you just use a single wide ‘two-shot’ throughout then the interviewer can provide unwelcome distraction as they are often only talking for a small portion of the time. In general you should avoid zooming in and out during the interview too much as this can be distracting and make it harder to encode.

After filming

Transferring footage from the camera to your computer for editing will involve connecting the camera directly with a USB cable, or removing the memory card and inserting into an external card reader or internal slot on your computer. It is a good idea to keep a back up of the media, making sure you have the video files in two places at all times – for instance on your computer and an external hard-drive. Make sure this is done before deleting the clips from your camera or memory card.

Some cameras record in formats that can be edited natively – in other words you can simply drag the clips from the memory card onto your computer and begin editing. Alternatively, the footage from some cameras will need to be ‘ingested’ into your computer – usually through a software application that comes with the camera. In the latter case you won’t be able to edit the files until they have been imported correctly.

There are numerous different types of editing software. At the basic end iMovie and Microsoft Photos - Video Editor are more than capable of producing very good edits with easy addition of text. At the higher end, software like Apple Final Cut and Adobe Premiere are very powerful creative tools. In general, you should accept the default video standard that your editing software suggests. There are many different formats but most modern software will simply optimize for the format you recorded in.

As well as cutting together the footage that you have shot, don’t forget that you can also add images, photos, on-screen text, PowerPoint slides and graphics to illustrate your video. If you are going to include graphics, make sure that they are optimized for the screen size of the video you are working with and also that fine detail isn’t going to disappear when they are encoded to a smaller screen size. For instance, while the small text on a graph might be very clear in your editing software, if you export a file that is 50% of the original screen size then the fine detail is probably not going to be visible.

You might also find that photos are in portrait rather than landscape format which corresponds to the screen format in wide-screen video. In this instance, don’t forget that you can zoom in on the detail on the photo making it fill the screen. This is generally much better than leaving black bars down each side of your image.

Exporting and encoding

Once your video is edited you will need to encode it for delivery within OUP’s digital platforms. In general, you will be editing in the format that your camera shoots. This will invariably mean that your video files are very large in data size. You should initially export a version of your video that is in the original format so you have a high-quality export for archiving. You can then either encode this export with separate encoding software, or export a lower-resolution version from your editing software. The reason for encoding is simply to reduce the file size so that it can be optimally uploaded and downloaded.

The main video delivery platforms for OUP are web delivery online and media embedded within PDFs. The ideal delivery format for both of these is a video codec called H264 or MP4. This is very widely used codec and produces very good quality video.

Your editing/encoding software might simply offer you the choice of small, medium and large encodes, or refer to a delivery destination – for instance ‘iPad’ or simply ‘web’. In this scenario you won’t have very much control over the encode parameters and should simply pick the one that is closest to the ideal.

As regards web delivery, OUP uses a video delivery system called Brightcove. This enables the upload of higher resolution videos which are then automatically encoded down into a variety of different encode sizes. This means that if someone is watching the video on a very good internet connection – for instance at a university – they will get served a very high resolution file. Conversely, if they are watching on a poor quality internet connection then they will get served a lower resolution version of the file.

As a result, it is preferable to supply a high resolution encode that can be uploaded by OUP to this system. You should therefore keep the original screen size and encode to H264. If you don’t have much control over the encode settings within your software then simply chose high-quality or YouTube or similar. If you do have control over the settings then the following would be appropriate for High Definition video:

Video Codec – H264 or MP4
Encode method – Multi-Pass
Aspect ratio – 16:9
Video size in pixels – 1920 x 1080
Frame rate – 1:1
Key Frame – Auto or 200 frames
Data rate – 3,000 Kbps

Audio codec – AAC
Audio data rate – 128Kbps
Channels – Stereo or mono
Sample Size - 16

If you have been filming in standard definition (for instance in a DV tape-based camera system) then the settings would be as follows:

Video Codec – H264 or MP4
Encode method – Multi-Pass
Aspect ratio – 16:9
Video size in pixels – 1024 x 576
Frame rate – 1:1
Key Frame – Auto or 200 frames
Data rate – 2,000 Kbps

Audio codec – AAC
Audio data rate – 128Kbps
Channels – Stereo or mono
Sample Size - 16

How to submit your files

Submit any .mp4 files to the journal alongside your manuscript file, using the naming convention Video_1.mp4, Video_2.mp4, Video_abstract.mp4 etc. Include a representative still image in .jpg format for each video, for use in the article PDF, using naming convetnion Video_1_still_image.jpg. If you encounter difficulties uploading especially large files to the journal's peer review system, please contact the journal's editorial office for guidance. Ideally you would send the following:

The original video export from your editing software without compression
A large sized encode in H264 format at the original screen size for upload by OUP to Brightcove for web delivery

Please note that the above filming tips constitute general guidance and are not meant to prescribe precisely what you should do. Successful filming is often a matter of experimentation to discover what works best in a particular scenario. We encourage you to explore the possibilities in this area as moving images can make a substantial contribution to the understanding of academic research.