Google Leverages AI for 3D Product Visualization in Shopping
Google is enhancing online shopping with new AI-powered technology. This innovative system simplifies 3D object capture, allowing businesses to showcase products in Google Shopping with interactive 3D models.
Google's new generative AI techniques create high-quality, shoppable 3D product visualizations using as few as three product images. This technology is already being implemented across various product categories on Google Shopping.
Leveraging Google's Veo AI video generation model, the system infers movement and fills in frames to create realistic 3D depictions. This surpasses previous 3D generation models in responsiveness and accuracy.
A key strength of Veo is its ability to generate videos that capture complex interactions between light, material, texture, and geometry. Its powerful diffusion-based architecture and its ability to be finetuned on a variety of multi-modal tasks enable it to excel at novel view synthesis.
This advancement allows brands to create compelling 3D product displays with minimal effort, improving the shopping experience for customers. The process also expands Google's 3D object library, potentially contributing to future VR shopping experiences.
We discovered that this approach generalized effectively across a diverse set of product categories, including furniture, apparel, electronics and more. Veo was not only able to generate novel views that adhered to the available product images, but it was also able to capture complex lighting and material interactions (i.e., shiny surfaces), something which was challenging for the first- and second-generation approaches.
While replicating the in-store experience perfectly remains a challenge, VR shopping enhanced with these 3D models could significantly bridge the gap. This innovative use of AI offers a valuable advancement for online retail.
Google is progressively rolling out these advanced object generation models for Google Shopping displays.
