When you anonymize data, you remove or change enough information in a data set to ensure that no single – or a combination – of values can be used to identify a person. This is useful when you need to share data, but do not have consent for sharing the identity.
There are multiple ways to anonymize data, and how you do it depends on your situation and field of research. In video you can do it by e.g. blurring the video, distort voices. Note that characteristics of a person may not only be the face, but can also be tattoos and the like, or a specific feature like a bag that is recognizable in a given context.
Be aware that some metadata might reveal information about your subject. This can be filenaming, but also embedded metadata like geographic location, custom metadata or similar that is not removed.