ZMO.ai gets Hillhouse-led $8M to create AI-generated fashion models – TechCrunch
With breakthroughs in machine learning, it is no longer uncommon to see algorithmically generated bodies that can move and speak authentically like real humans. The question now is whether startups offering such products can achieve a sustainable business model. Some of them have demonstrated this potential and attracted investors.
ZMO.ai, founded by a team of Chinese entrepreneurs who have spent years studying and working overseas, has just closed an $8 million Series A funding round led by Hillhouse Capital. GGV Capital and GSR Ventures also participated in the funding round.
The startup has found healthy demand from fashion e-commerce companies who struggle to hire and afford models due to their growing number of stocking units (SKUs) or styles, as consumer tastes are becoming more changeable. Using the Generative Adversarial Network (GAN), ZMO created software to help them create virtual full bodies of models by setting simple parameters like face, height, skin color, body shape, and laid.
“Traditionally, the full garment manufacturing cycle can take two to three months, from design, fabric selection, pattern making, modeling, to shelf launch,” says Ella Zhang, CEO and co-founder of ZMO, former engineer at Google. and Apple.
“We are reversing and shortening this process. [Customers] can now test a garment by putting it on a virtual model, which can go on the site. Once orders are received, the e-commerce client can begin manufacturing,” she told TechCrunch. “They can also test what kind of people would suit a certain product by trying it on different virtual models.”
It’s no surprise that fashion e-commerce operators see ZMO and its ilk as a cost-cutting tool. Zhang says her company is in talks with fast fashion giant Shein, which launches 2,000 to 3,000 new products a day, about potential collaborations.
We’ve already covered Surreal, a Shenzhen-based startup backed by Sequoia, which is also working on synthetic media to replace humans in lifestyle photos and other business scenarios. The company sparked renewed interest when the COVID-19 pandemic hit Chinese e-commerce exporters, who struggled to find overseas models as the country entered tight border controls.
In the future, ZMO also plans to apply GPT-3, which uses big data and deep learning to mimic the natural language models of humans, to create speech for the models. Scary as it sounds, this feature would allow e-commerce businesses to quickly and cheaply produce TikTok videos for product promotion.
On average, e-commerce businesses spend about 3-5% of their annual gross merchandise value (a rough metric measuring sales, usually excluding returns and refunds) on photoshoots, according to Roger Yin, who worked at Evernote. and ran his own cross-border e-commerce business before co-founding ZMO with Zhang.
“Images play an important role in driving online sales. The problem is that the [sales] The cycle is short but the cost of images is high,” observes Yin, adding that costs can be even higher for fashion companies with rapid style turnover. ZMO’s goal is to reduce photoshoot costs to 1% of GMV.
At present, 80% of ZMO’s customers are based in China, but it is trying to attract more overseas users this year with its new financial injection. Operating with a team of 30, the startup has 30 “medium and large” customers, including Tencent-backed Chicv, one of Shein’s many challengers, and more than 100 “small and medium” customers, such as dropshipping sellers. .
ZMO’s other co-founders include Ma Liqian, a Ph.D. in computer vision with a degree from KU Leuven in Belgium, and Yang Han, who previously worked on AI-based style at Tencent and SenseTime.