Here are some ‘PrinGital’ – print and digital – startups from India disrupting the visual recognition industry
Shikha Gupta came across a pretty dress in a newspaper advertisement of a local fashion store. She knew it would look great on her. But Gupta didn’t want to take the trouble of visiting the shop to buy it.
For the brand, that’s a broken sales process. Though ROI (Return on Investment) driven advertising tops the list of brand marketers, tools to optimize print ads, which attract 41 percent of total ad spending, didn’t exist until recently. However, some Indian startups are experimenting with apps that make advertisements instantly shoppable.
As these startups mature, brands will get to leverage the power of the moment when users are already engaged on a print ad. They will be able to convert customer impulse into a sale.
Like a virtual salesman, these startups are also building features that recommend similar products. These transform static print advertisements into something engaging and transactional.
Here are some ‘PrinGital’ – print and digital – tools from India.
When users view a Voconow-powered print ad through the app, it instantaneously recognizes the ad and overlays a digital layer on top of it. On the app, users can see the products from different angles, buy the products, send an inquiry, book a test drive in the case of a car, schedule a site visit of a real estate project, and so on.
In addition, Voconow provides analytics on the ad’s customer engagement. It gives insights into the demographics of engaged customers for marketing decisions like designing future product promotion campaigns, spotting trends, consumer behavior, and so on.
Voconow also serves as a channel to order complimentary products.
Voconow’s first customer was Lenovo, and together they produced interactive ads that allow users to see the Vibe X2 smartphone with different colors, view product specs, and buy the smartphone right from the print ad. Voconow declined to comment on the number of smartphones sold through its platform.
The startup has been selected for Facebook’s startup mentorship program FBStart.
Another startup in this space is Wazzat. It was born out of researcher Jay Guru Panda’s work on image recognition while he was pursuing his MS degree at IIIT-Hyderabad. When he applied the research to solve real world problems, he realized that fashion e-commerce is the segment in need of it. And so Wazzat was born. With its flagship product Wazzat Fashion, consumers can download a partner retailer’s app or snap a photo of an apparel and instantly search for similar clothes.
Unlike Voconow, Wazzat developed APIs that can be integrated into existing apps of retailers. It also has social media plugins for searching similar products online when users notice them on Pinterest or Instagram.
It has proprietary auto-tagging tools to ensure standardization and accuracy which the “prevalent online stores lack,” says Wazzat founder Mauktik Kulkarni. He explains, “The technology used for product recommendations is collaborative filtering. This means if you bought a beautiful summer dress and bought a book with it, the next customer buying the same dress will see a book in her recommendations.”
Wazzat claims to differentiate itself by focusing on the style of the garment and not just color and design pattern. For example, it can distinguish between an evening gown, floor-length dress, spaghetti straps, or a strapless dress. “Horizontal players do not have the bandwidth to look into these details and develop expertise for it in-house. Our specialized technology works in cluttered backgrounds too,” says Mauktik.
Wazzat Fashion is currently live on U.S.-based retailer Target’s website. It is also in the pilot stage on a couple of Indian e-commerce stores.
iLenze is another B2B smart visual recognition and visual-based product search company that claims to have integrated with online furniture and fashion stores, as well as online classifieds portals. The startup claims to be more efficient than others by using “machine learning, artificial neural networks, and deep learning combined with parallel computing to process and understand images how a human eye does.”
“The technology enables us to solve complexities like identifying the target objects in images with cluttered backgrounds, varying lighting conditions, and varying object distances while ignoring what’s not contextual. It can process a huge amount of visual data in real time,” said iLenze founder Ashish Kumar, who ran fashion aggregator Fashupp before pivoting.
According to Kumar, visual recognition has relevance not just in fashion but also medicine. For instance, scanning x-ray reports can find tumors and detect the intensity of a disease.
The startup is being incubated at Times Internet’s startup incubator TLabs and has raised an undisclosed amount in angel funding.
Artificial intelligence (AI) researcher Anand Chandrasekaran left the Neurogrid project, which was building a device that simulates the neurons and synapses of the brain to control bionic limbs, to experiment with AI in other spaces. He got the smartphone camera to detect facial expressions and respond appropriately. For example, showing a frown to the camera could stop an unwanted call.
Today, that has evolved into a full-fledged company – Mad Street Den. Its flagship product is MadStack, which does visual search for fashion items using its object recognition module. It also offers gaze tracking, emotion-expression detection, head and facial gestures, and 3D facial reconstruction capability. According to the company, all these have applications in fashion, gaming, internet of things, robotics, automotive, and analytics.
Mad Street Den raised $1.5 million in seed funding from Exfinity Fund and GrowX Ventures earlier this year.
Editing by Terence Lee
This post was originally published on Tech in Asia.