Arif Mahmood

Title: Arif Mahmood and Label-Preserving Data Augmentation

Introduction

Arif Mahmood is an inventor associated with Lahore, Pakistan, and Mohamed Bin Zayed University of Artificial Intelligence. His patent work focuses on using diffusion models to create enriched training images while preserving image labels.

Latest Patents

Mahmood is associated with 1 patent: “System and Method for Label-Preserving Data Augmentation with Diffusion Models.” The invention describes a method for generating augmented training images from an original image and one or more prompts supplied to a diffusion model.

The method combines part of the original image with part of the generated image using a binary pixel-wise mask. It then blends a fractal image into the hybrid image to produce the augmented result. The approach is designed to enrich training data while retaining label-relevant content.

Career Highlights

Mahmood’s listed patent reflects work at the intersection of artificial intelligence, image generation, and data augmentation. His affiliation with Mohamed Bin Zayed University of Artificial Intelligence connects his work to an institution focused on artificial intelligence research and development.

Collaborations

Khawar Islam and Muhammad Zaigham Zaheer are identified as Mahmood’s coworkers. Their shared professional context is associated with the university and the research area represented by the patent.

Conclusion

Arif Mahmood’s patent describes a diffusion-based image augmentation technique that combines original, generated, and fractal imagery to support enriched training data.