Composer by Advex provides an on-device generative AI platform for automated visual inspection tasks. This system supports multi-class classification and semantic segmentation to accurately identify and locate defects on manufacturing lines. Its language-driven UI allows for rapid deployment and integration without requiring specialized AI expertise.
Funding
$3.5M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Computer vision models require extensive labeled data for training, which is often time-consuming and expensive to collect, especially for edge cases or variations that are difficult to capture in real-world scenarios. The lack of sufficient data can lead to reduced model accuracy and adaptability, hindering the deployment of effective vision AI solutions.
Solution
Advex AI offers a Vision AI platform that leverages generative AI to create synthetic data for training and enhancing computer vision models. By generating thousands of labeled images from a small set of real-world images, the platform significantly reduces the time and cost associated with data collection and labeling. The generated synthetic data addresses data gaps and improves model performance, enabling faster deployment and adaptation to new products, defects, and process changes. The platform identifies key gaps in existing datasets and automatically generates the necessary synthetic images to fill those gaps, resulting in higher accuracy and more robust computer vision models.
Target Audience
The primary target audience includes companies in logistics, manufacturing, agricultural robotics, and retail that utilize computer vision for tasks such as object detection, defect detection, and robotic guidance.
Features
- Generates thousands of synthetic images from as few as 5-10 real images
- Identifies and addresses data gaps in existing datasets to improve model accuracy
- Supports rapid adaptation to new products, defects, and process changes through on-the-fly data generation
- Provides labeled synthetic data, eliminating the need for manual labeling
- Enables the creation of diverse datasets with variations in lighting, weather conditions, and object characteristics
- Offers case studies demonstrating improved accuracy in logistics, manufacturing, and other applications