HomeArtificial IntelligencePresti replaces expensive photo shoots within the furniture industry with GenAI

Presti replaces expensive photo shoots within the furniture industry with GenAI

If you've ever bought a settee online, have you ever thought in regards to the houses you possibly can see within the background of the product photos? When it's time to launch a brand new collection, furniture brands often spend a small fortune on photoshoots. It's a laborious and expensive process, because it's hard to maneuver furniture around.

That is why a French startup called priestfounded in November 2022, uses generative AI to show a single product image into a practical lifestyle photo. The company just closed a $3.5 million seed funding round led by global tech investment firm Partech, with participation from several business angels.

“We in a short time picked up the phone and spoke to 50 potential users,” Nabil Toumi, co-founder and CEO of Presti, told TechCrunch. “And all of them said the identical thing. Creating product images was a really time-consuming and dear process, and there was no easy solution to create these photos. At the identical time, it was an important asset for brands to construct a novel identity and sell online.”

Presti turns it into:

Photo credits: priest

This could also be:

Photo credits: priest

Initially, Presti didn’t limit its focus to furniture corporations. But the startup quickly realized that furniture corporations faced some particularly difficult problems.

“For a photograph shoot, they’d to rent a good looking house and transport products – this resulted in high logistics costs. And these photo shoots were planned months prematurely and ended up costing them tons of of 1000’s, even thousands and thousands of euros per yr,” said Toumi.

At its core, Presti uses Stable Diffusion XL as its base model. It has been retrained and optimized to be particularly well suited to product images within the furniture industry.

At first, the team tried using an ordinary version of Stable Diffusion XL, nevertheless it quickly became clear that there have been problems. “Your sofa gets legs and the backrest gets distorted,” said Toumi. Getting the angle right was also difficult. For example, the wall behind the sofa needs to be parallel to the sofa.

“At the identical time, the dataset we used to coach our model is basically essential. We currently have greater than 75,000 ultra-high quality images of furniture in our industry that we are able to use to coach our model to bolster the educational process for a selected use case, for that style of photo,” said Toumi.

Presti didn't need to limit itself to background generation. Customers may also add accessories. For example, for those who're creating product photos for a brand new sofa, you possibly can add pillows. These pillows will solid a practical shadow on the sofa, so that they don't appear like they were added in Photoshop.

Likewise, furniture brands often offer multiple variations of the identical model with different textures and colours. While this remains to be a piece in progress, Presti hopes its customers will find a way to swap out the fabric using its tool. This will make it much easier for corporations using Presti to launch latest products.

On the opposite hand, freelance photographers won’t like this latest product. And whether the creativity and originality that other expert people working on a physical photoshoot, resembling location stylists, bring to the table will be completely replaced by machine-generated backgrounds – without the resulting artificial lifestyle images looking just a little samey – is an open query.

Although Presti mainly works with medium-sized furniture corporations, it also has a strategic partnership with Maisons du Monde, one among France's largest furniture retailers. In addition to Partech, the startup's other investors include Maxime Brousse, Thibaud Elzière, Julien Hirth, Abou Laraki and Rémi Lemonnier.

LEAVE A REPLY

Please enter your comment!
Please enter your name here

Must Read