Interesting AI Projects Built using Intel® oneAPI Tools

Highlights of Intel’s AI Hackathons at Daksh and Anokha tech events

Intel recently conducted two AI hackathons (with a major focus on generative AI, or GenAI) in India. Participants developed innovative AI projects using Intel® oneAPI tools and the Intel® Tiber™ Developer Cloud platform. One of them was part of the Daksh 2024 tech fest organized by SASTRA University on March 8, and the other was part of the Anokha 2024 tech event organized by Amrita Vishwa Vidyapeetham on April 19.

This blog will give you an overview of both the Intel AI hackathons (organized by Intel’s Ecosystem Developer Program team) and amazing projects developed at the events.

Hackathons At A Glance

• Around 1200+ developers (students, professionals, and other tech enthusiasts) attended the Anokha hackathon while 550+ collegiate developers participated in the Daksh hackathon.
• 600+ users at the Anokha hackathon and 300+ users at the Daksh hackathon newly signed up at Intel Tiber Developer Cloud.
• Through 30 mentoring sessions at the Anokha hackathon and 3 workshops at the Daksh hackathon, the participants got a deep dive into Intel Tiber Developer Cloud and Intel’s resources for AI/ML and GenAI.

Top AI Projects at the Daksh Hackathon

Out of the 27 finalist projects, the following top 4 won the GenAI focused Anokha hackathon:

1.  Stylist AI, the 1st prize winner, is a generative AI-based solution for personalized outfit recommendations that can improve user experience on online e-commerce websites. The team employed Intel® Extension for PyTorch*, Intel® Extension for TensorFlow*, Intel® oneAPI Deep Neural Network (oneDNN) library, and Intel Tiber Developer Cloud.
   
   
2.  Llama Hunt, a 2nd prize winner, is a Large Language Model (LLM)-based web application that helps find job openings tailored to the applicant’s resumes and preferences. The team utilized Intel Extension for PyTorch for faster inference and Intel Tiber Developer Cloud for accelerated deployment of the project.
   
   
3. Find My Doctor, also a 2nd prize winner, is a mobile application designed to help users find the best doctors available near them based on symptom analysis and patients’ and hospitals’ classification. It thus ensures prompt access to healthcare. The team leveraged oneAPI’s optimizations of machine learning libraries and Intel Tiber Developer Cloud for faster execution on GPUs.
   
   
4. Emerge AI, the 3rd prize winner, is a call management solution designed to enable timely responses to critical situations for efficient emergency services. Using optimized ML and Natural Language Processing (NLP) frameworks and algorithms of oneAPI, it prioritizes emergency calls in the absence of call handlers by tracking the caller’s location and analyzing the call's keywords.

Winner AI Projects at the Anokha Hackathon

Out of the 11 finalists, following were the winning AI projects at Daksh hackathon that made the best use of Intel’s AI software stack:

1. Comic-ify, the 1st-prize winner, is an LLM-based application that converts mundane PDFs into visually engaging comic-style content for a better reading experience. Its fine-tuned GenAI model generates imaginative text and images based on text. The team leveraged Intel Tiber Developer Cloud platform and optimized Python* libraries provided by Intel® Distribution for Python*.
   
   
2.  DeepfakeDetection, the 2nd prize winner, is a solution that employs advanced ML models for detecting and mitigating deepfake media through image classification, video classification, audio analysis and a combination of these techniques. The AI frameworks used in the project include Intel Extension for PyTorch, TensorFlow* optimizations from Intel and oneDNN. The team also used accelerated hardware resources available on Intel Tiber Developer Cloud.
   
   
3.  Traffic-signal-AI is a project that aims at dynamically adjusting traffic signals to prioritize passage of emergency vehicles while effectively managing the traffic flow. It leverages Intel Extension for PyTorch, Intel Extension of TensorFlow, and high-speed CPUs on Intel Tiber Developer Cloud for real-time video analysis, audio processing, and optimized ML algorithms. AI Tools from Intel help perform faster data processing, model training, and inference in the project.
   

What’s Next?

Get started with Intel Tiber Developer Cloud for accelerated AI development with optimized oneAPI tools and software optimizations on the latest Intel hardware. If you are new to the cloud platform, sign up today!

We also encourage you to learn about other AI, HPC, and Rendering tools in Intel’s oneAPI-powered software portfolio.

Additional Resources

• Intel’s AI/ML Ecosystem Developer Resources
• AI Tools from Intel
• Intel’s GenAI resources
• Intel’s AI/ML development tools and resources
• Intel’s AI/ML frameworks optimizations
• Intel’s AI hardware solutions