Koraiio

Research & publications.

The published work behind what we build. Most of it is the founder's research from graduate school, industry and Regis University, which Koraiio now builds on. Every item links to its DOI, patent record or code.

Publications

2026

  1. Real Time Pixel Level Esports Prediction Using Edge AI Gregory, T., & Mujtaba, G. 54th Applied Imagery Pattern Recognition Workshop (AIPR), Lecture Notes in Computer Science, Springer Computer vision

2025

  1. Figure from EdgeVidSum: Real-Time Personalized Video Summarization at the Edge
    EdgeVidSum: Real-Time Personalized Video Summarization at the Edge Mujtaba, G., & Ryu, E. S. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Demo track Edge AIDOI 10.48550/arXiv.2506.03171Project page
  2. Figure from EdgeAIGuard: Agentic LLMs for Minor Protection in Digital Spaces
    EdgeAIGuard: Agentic LLMs for Minor Protection in Digital Spaces Mujtaba, G., Khowaja, S. A., & Dev, K. IEEE Internet of Things Journal Language modelsDOI 10.1109/JIOT.2025.3574961Project page

2024

  1. FISTNet: Fusion of Style-Path Generative Networks for Facial Style Transfer Khowaja, S. A., Nkenyereye, L., Mujtaba, G., Lee, I. H., Fortino, G., & Dev, K. Information Fusion Computer visionDOI 10.1016/j.inffus.2024.102572

2023

  1. Figure from FRC-GIF: Frame Ranking-Based Personalized Artistic Media Generation Method for Resource Constrained Devices
    FRC-GIF: Frame Ranking-Based Personalized Artistic Media Generation Method for Resource Constrained Devices Mujtaba, G., Khowaja, S. A., Jarwar, M. A., Choi, J., & Ryu, E. S. IEEE Transactions on Big Data Video & multimediaDOI 10.1109/TBDATA.2023.3338012CodeProject page
  2. Education 5.0: Requirements, Enabling Technologies, and Future Directions Ahmad, S., Umirzakova, S., Mujtaba, G., Amin, M. S., & Whangbo, T. arXiv preprint arXiv:2307.15846 OtherDOI 10.48550/arXiv.2307.15846

2022

  1. Figure from LTC-SUM: Lightweight Client-Driven Personalized Video Summarization Framework Using 2D CNN
    LTC-SUM: Lightweight Client-Driven Personalized Video Summarization Framework Using 2D CNN Mujtaba, G., Malik, A., & Ryu, E. S. IEEE Access, vol. 10, pp. 103041–103055 Computer visionDOI 10.1109/ACCESS.2022.3209275CodeProject page
  2. Figure from Client-Driven Lightweight Method to Generate Artistic Media for Feature-Length Sports Videos
    Client-Driven Lightweight Method to Generate Artistic Media for Feature-Length Sports Videos Mujtaba, G., Choi, J., & Ryu, E. S. 19th International Conference on Signal Processing and Multimedia Applications (SIGMAP), pp. 102–111 Video & multimediaDOI 10.5220/0011335300003289Project page

2021

  1. Figure from Client-Driven Animated GIF Generation Framework Using an Acoustic Feature
    Client-Driven Animated GIF Generation Framework Using an Acoustic Feature Mujtaba, G., Lee, S., Kim, J., & Ryu, E. S. Multimedia Tools and Applications, 80(10) Acoustic sensingDOI 10.1007/s11042-020-10236-6CodeProject page
  2. Human Character-Oriented Animated GIF Generation Framework Mujtaba, G., & Ryu, E. S. Mohammad Ali Jinnah University International Conference on Computing (MAJICC), pp. 1–6 Computer visionDOI 10.1109/MAJICC53071.2021.9526249

2020

  1. Figure from Client-Driven Personalized Trailer Framework Using Thumbnail Containers
    Client-Driven Personalized Trailer Framework Using Thumbnail Containers Mujtaba, G., & Ryu, E. S. IEEE Access, vol. 8, pp. 60417–60427 Video & multimediaDOI 10.1109/ACCESS.2020.2982992Project page

2019

  1. Energy Efficient Data Encryption Techniques in Smartphones Mujtaba, G., Tahir, M., & Soomro, M. H. Wireless Personal Communications, 106(4), pp. 2023–2035 OtherDOI 10.1007/s11277-018-5920-1Code

Full list on Google Scholar ↗

Patents

Card for U.S. provisional patent application 64/148,522, inventor Ghulam Mujtaba, Koraiio LLC

U.S. provisional patent application, 2026

Passive acoustic detection, classification, localization, and tiered defeat cueing of unmanned aerial systems

Koraiio LLC (DroSonic). Inventor: Ghulam Mujtaba

US 64/148,522

Patent drawing: a Recommended for You screen offering summary types 1 to 3 or No Thanks

U.S. patent application publication, 2024

Personalized semantic fast-forward videos for next generation streaming platforms

Inventors: G. Mujtaba, E. S. Ryu, R. Harb

US 2024/0073493 A1 ↗

Open-source code

Reference implementations published with the papers, and tools used along the way.

Talks and demos

  1. Keynote, 54th Applied Imagery Pattern Recognition (AIPR) WorkshopThe George Washington University, Washington, D.C.
  2. DroSonic: Passive Acoustic Detection and Cueing of Fiber-Optic and Radio-Silent UAS on Low-Cost Edge AI HardwareNDIA Emerging Technologies for Defense Conference, Washington, D.C.
  3. Keynote: Delivering Personalized Content on Edge AIInternational Conference on Computing & Innovation (ICCI-2025), Lahore
  4. EdgeVidSum live demoCVPR 2025, Nashville, Tennessee

Hands-on writing

Step-by-step technical notes on the founder's site, mostly about getting models and video running on small hardware.

  1. DeepSeek-R1 WebUI on an NVIDIA Jetson with Ollama
  2. DeepSeek-R1 running in the browser with WebGPU
  3. A multimodal web interface for LLaMA 3.2 with Ollama
  4. Generating thumbnail containers from video
  5. Configuring an HLS streaming server
  6. Hardware-accelerated FFmpeg on the NVIDIA Jetson TX2

All posts on gmujtaba.com ↗

DroSonic OSINT

Our weekly open-source assessments of drone and counter-UAS developments: what flew, what stopped it, and what it means for detection.

  1. Hearing the Drop
  2. When Jamming Stops Working
  3. Watching the Bases
  4. Thirty-Four Thousand Flights
  5. Allowed to Listen, Not to Shoot
  6. Closed on a Sighting

All 16 assessments ↗

Have a perception problem that has to run on real hardware?

Tell us what you need to detect, where it has to run and what it runs on. You'll get a reply from the founder.

Email gmujtaba@koraiio.com

Email
gmujtaba@koraiio.com
Founder
Dr. Ghulam Mujtaba
Product
DroSonic
Based in
Denver, Colorado