2024-09-11 05:23:51
Saturday, September 07, 2024 – 13:40 WIB
Google Shopping illustration. Photo: The Guardian
jpnn.comJAKARTA – Google is increasing the use of generative artificial intelligence, which allows its users, especially women, to virtually try on thousands of dresses from hundreds of brands.
According to Google, dresses are the most searched clothing category using the try-on tool.
However, existing diffusion techniques are difficult to use on dresses, which are more detailed and complex than other types of clothing.
Google Shopping last year launched a clothing try-on tool that uses diffusion technology to create high-quality, life-like images of tops and blouses.
The tool simulates how clothing will fall, fold, stick, and form wrinkles and shadows on people in various poses.
Due to the intricate details on the dress, existing diffusion models have difficulty capturing detailed prints such as floral or geometric patterns.
Therefore, Google developed a new training strategy that starts with low-resolution images, and gradually incorporates higher resolutions.
Additionally, the company is using new techniques to deliver a more accurate depiction of the dress on people.
Google is ramping up its use of generative artificial intelligence (AI) to virtually try on dresses
Please read other interesting content from JPNN.com on Google News