Data Science

Winners Announced From NVIDIA Inception Program Contest

AI-Generated Summary

  • The NVIDIA Inception Program hosted a Cool Demo Contest for GPU-accelerated startups applying deep learning to innovations across defense, healthcare and other fields.
  • SmartPlate TopView by Fitly uses TITAN X Pascal GPUs and the Caffe deep learning framework to instantly analyze and track over 1,000 food types for nutrition monitoring.
  • Imagia leverages cuDNN and NVIDIA GPUs to provide AI-based predictive oncology analytics on medical imaging data, giving radiologists a second opinion on cancer detection.
  • TLD Vision developed xTLD with CUDA, TITAN X GPUs and cuDNN for real-time facial detection and 360-degree pose estimation in HD video, with plans to extend the technology to object tracking for drone-based coastline protection and power-line inspections.

Next Steps

Powered by NVIDIA Nemotron. AI-generated content may summarize information incompletely. Verify important information. Learn more

Late last year, the NVIDIA Inception Program hosted a “Cool Demo Contest” for GPU-accelerated startups that are applying deep learning to their innovations.
A variety of companies from around the world submitted their demos, ranging from defense to healthcare applications. Below are highlights from three of the 14 winners who each won a Pascal TITAN X GPU.
Eating Smart Just Got Smart
Now available on Indiegogo, SmartPlate TopView by Fitly is the world’s first Intelligent Nutrition Platform that instantly analyzes everything you eat. Instead of manually logging your food items, SmartPlate’s AI technology is able to analyze, and track everything you eat in seconds. With the help of TITAN X Pascal GPUs and Caffe deep learning framework, the product can successfully detect over 1,000 different types of food.

AI for Medical Image Analysis
Montreal-based Imagia, founded by serial entrepreneurs, provides AI-based actionable predictive oncology analytics atop medical imaging data to give radiologists an accurate and timely second opinion on cancer detection. The startup is using cuDNN and NVIDIA GPUs for both training their deep learning models and for classifying possible types of cancer.

Real-Time Detection and Alignment of 3D Objects

TLD Vision, located in the Czech Republic, developed a new product codenamed ‘xTLD’ that uses CUDA, TITAN X GPUs and cuDNN to do real-time facial detection and 360° pose estimation in HD video. The company plans to adapt their recognition technology for different objects – and not just faces, such as tracking a shark from a drone for coastline protection, or detection of poles for power-line inspections.

The other demo contest winners were Aetros, Airxsys, Colorful Clouds, CTAAdventure, Iceberg Hockey Analytics, Imagr, Malong, NLP Logix, and Understand.AI.
Is your startup revolutionizing the retail industry using artificial intelligence and GPUs? If so, submit your application for the Inception Program Retail Contest by February 24, 2017.

Discuss (0)

Tags