The Stack Archive

Coming soon: movie trailers that change depending on who you are

Fri 18 Mar 2016

A Cambridge academic is working on a project to auto-generate movie trailers based on the user watching them, emphasising aspects of the movie likely to appeal most to that viewer.

Among a fascinating raft of projects based on computer-recognition of visual data, Shasha Lu, a lecturer in Marketing at Cambridge Judge Business School, proposes that the dark art of movie trailer production could become automated to produce bespoke output, and explains:

“[As] it stands all move trailers target the same group of people; if there’s, say, a trailer with an action element that’s watched by someone that doesn’t go in for action films, they may well be put off seeing it, whereas in fact the movie might have plenty of non-action that makes the film attractive to this viewer.”

The field of preference-targeted trailers has so far existed only as a meme on YouTube, where users upload spoof trailers that defy the genre of the movie in question, such as this popular one which attempts to sell Stanley Kubrick’s 1980 horror classic The Shining as a genial family comedy:

Marketing a ‘Four-Quadrant’ movie – one which holds some appeal both for men and women and for under and over-25s – can be problematic, leading to enormous variation in types of publicity that are run for the film, depending on the context (such as men’s vs. women’s magazines, and the target audience, hour and demographics of a TV ad). But a trailer delivered via a website to a computer or smartphone could potentially be chosen from a pool of possible demographically-targeted versions – or even assembled ad hoc. Since the project is currently at an experimental stage, Lu gives no details about the mechanism or viewing environment considered, but since most of her projects involve either facial analysis or other kinds of intelligent visual analysis of consumers, these could be input factors.

“We know that it’s possible to analyse the content of movies to extract all kinds of information, “ says Lu. “Lighting, colour, motion and shot, for instance – all of which relates to people’s preferences, and we are working on using this to customise movie trailers according to people’s preferences, thereby improving the effectiveness of the movie trailer and getting people to watch movies they might otherwise dismiss.”

Lu, who was born in Shanghai and holds degrees in computer science and business, together with a PhD in marketing, is also working on an automated recommender system to help women choose clothes in stores, which uses a camera mounted on a mirror to analyse the woman’s facial expressions and behavioural responses, two pieces of ‘data’ which are of great interest to the sales people attending them: “The first is whether the customer likes it or not, which is inferred by her emotional response from her facial expression. The second is which particular part of the garment she likes or dislikes, which is observed through behavioural response. If she’s touching her collar or scratching the lower hem with an annoying expression on her face, for example, we may infer she doesn’t like that bit of the garment.”

Lu, who is also developing a dating recommending system based on people’s preferences for certain facial characteristics and topography, considers that the field of image and video data analysis is going to be of increasing interest in marketing in the future, and notes that Harvard Business School has already taken an avid interest in the potential of such technologies.


advertising news research UK
Send us a correction about this article Send us a news tip