Case Study: Teezily achieves automated image quality control with Addepto’s AI image detection engine

Teezily automates image quality checks with Addepto to catch bad uploads before production

Teezily, a global eCommerce platform, faced significant challenges with the quality of images uploaded by its users, which impacted final product quality and customer satisfaction. They needed to automate image quality assessment to scale their operations but lacked the in-house AI expertise. They partnered with the AI consulting firm Addepto to develop a computer vision solution.

The Addepto team built an automated image quality detection engine using the VGG convolutional neural network, trained on Teezily's own user-uploaded data. The solution was delivered as a ready-to-integrate software module that seamlessly connected with Teezily's backend. By implementing this system, Addepto enabled Teezily to automatically flag low-quality uploads for issues like blurriness and poor cropping, preventing them from entering production and thereby maintaining high product standards efficiently.

Teezily