AI Trends in Intelligent Transport Systems
summary
History of vision models leads to concrete claims: Navya runs classical image‑processing for road marking and obstacle detection and is porting deep networks to Nvidia automotive platforms, not just demos with a GPU in the trunk. Annotation comes from crowdsourcing and services like Amazon Mechanical Turk, and the field scales with annotated data volume. Demos (stereo from one camera, end‑to‑end steering) exist but production needs automotive‑grade GPUs and certification and multi‑process scene understanding beyond simple object labels.
AI-written summary · about
chapters (8)
- 0:00Intro
- 0:53ARTIFICIAL NEURON: A 80 YEARS OLD FELLOW
- 2:08PERCEPTRON - ROSENBLATT 1957: A NEW PARADIGM
- 5:07NEURAL NETWORKS - APPLICATIONS
- 8:30COMPUTING POWER: THE RISE OF GPU'S
- 11:07TRANING BASE: NEW NEEDS LEAD TO NEW BUSINESS MODELS
- 11:20LEARNING BASE: NEW NEEDS LEAD TO NEW BUSINESS MODELS
- 13:23DEEP LEARNING : APPLICATIONS
related talks
42mAI and Self-Driving Cars
05mInterview
28mMachine Learning Inside the Car
31mDeep learning for self-driving vehicle technology
52mPhysical AI: Crafting Resilient Systems with Emotional Intelligence
20mComputer Vision at the Edge with OpenVINO
19mHow to Deploy Artificial Intelligence in Industrial Application
24mAI in maritime - Autonomous ships