Marketers of the 1960s and 1970s would vouch for the efficacy of sending out surveys and paying a premium for prime-time brand placements as a part of TV advertising. With the current digital age, these traditional methods have given way to more targeted and contextual marketing methods.
Today’s digital marketing ecosystem is multi-channel, device agnostic, multi-platform, and driven by programmatic ads. Studies show that 80% of the internet traffic will be driven by video by 2019. With the online marketing world increasingly driven by video content, there emerges a need to place emphasis on making sense of the exploding visual data.
This is precisely where computer vision and machine learning come in.
Applications of computer vision?
Computer vision employs AI and big data to drive various marketing objectives like enhancing programmatic video capabilities, unraveling semantics in social media posts and images. Want to check out how this is done in the real world? Then check out a few use cases to understand the prowess of computer vision for marketing
1. Smart e-retail
Merchandising places a focus on tagging to attach attributes that users can use to find specific product choices. However, solutions like Sentient Aware goes one step ahead and facilitates visual product discovery rather than relying on just text. With this standard, e-commerce shoppers can select visually similar products to the one they are looking for or have browsed for in the past.
A popular sunglass brand Sunglass Hut uses this tech to refine the range of products displayed online based on visual similarities. The benefit is that it expands the range of products that are displayed to the shoppers and carried a better likelihood of purchase because of visual similarity
2. Contextual image display
The in-image ads are a great way to engage with site viewers. For example, a luxury auto forum may place a contextual ad for a new BMW car launch. This is enabled by the context capture capabilities of brands like GumGum that use in-image ads to ensure better engagement and higher conversions.
The underlying principle is it identifies what is in the image (for example a cat photo, or a car photo) and then places a relevant ad on the image itself. This is done by computer vision and improved ability to understand semantics (human understanding of images)
3. Emotional analytics
The recent $33billion company, Mediacom tie-up with emotion analytics company Realeyes has grabbed headlines in the marketing world. The analytics solution from Realeyes will be integrated with Mediacom’s central content hub. The key idea here is to get behavioral information. This is achieved by placing webcams in remote panels of users viewing the ads or content generated by Mediacom. This helps them track the facial expression and develop emotional analytics.
These examples show the immense potential for employing computer vision and machine learning to improve the efficacy of targeted marketing message and drive better conversions for a particular brand.