RealitySummary System Architecture

RealitySummary: Exploring On-Demand Mixed Reality Text Summarization and Question Answering using Large Language Models

RealitySummary integrates OCR, markerless tracking, and GPT-based models to provide on-demand document augmentation. Iteratively developed over three studies—usability, in-the-wild, and diary—it demonstrates significant potential for real-world reading enhancement through MR-AI integration.

June 12120 · Tafreed Ahmad
YOLOv8 Model in Action

The YOLOv8 Edge: Harnessing Custom Datasets for Superior Real-time Detection

The paper details a YOLOv8 model trained on custom datasets, achieving a mAP50 of 0.864 and a mAP50-95 of 0.758 for detecting objects in real-time streams, demonstrating advancements in accuracy and speed.

June 7070 · Tafreed Ahmad, Ahmad Maaz, Danyaal Mahmood, Zain ul Abideen, Usama Arshad, Raja Hashim Ali
Siamese Neural Network in Action

Influencing Factors in Facial Recognition and Verification: A Siamese Neural Network Hyperparameter Study

The paper examines the effect of hyperparameters on Siamese neural network performance in facial verification tasks. It emphasizes the significance of diverse datasets and optimal hyperparameter selection to enhance model accuracy.

June 23230 · Ahmad Maaz, Tafreed Ahmad, Shaheer Abbas, Usama Arshad