< Back

Introducing The Nexar Challenge: Deep-Driving into the Future

Introducing The Nexar Challenge: Deep-Driving into the Future

Use Deep Learning for Traffic Light Recognition, compete to win prizes and join our mission to make the road safer



Nexar is building the first over-the-top vehicle-to-vehicle (V2V) network app that turns smartphones into connected AI dashcams. Leveraging millions of crowdsourced road miles from dozens of countries, jointed with deep learning on imagery and sensors as well as artificial intelligence driving-cognition technologies, Nexar provides a new, delightful, and really safe driving experience. This technology has the potential of saving the 1.3 million people who die on the road every year.

Since launching in February 2016, we have tracked upwards of 20 million miles and recorded more than a half-million instances of driving incidents worldwide — and we’re just getting started.

Although remarkable progress has been made in computer vision, the vast majority of both existing theories and technologies have yet to transition to the real automotive world, where there is a huge variety of road infrastructure, side buildings, road signs, vehicles, and most importantly, human driving behaviors.

At Nexar, we’ve assembled a first class technical team dedicated to infusing state-of-the-art deep learning techniques into our world to help make it collision-free. Deep learning plays a key role in many of our ongoing projects: from a real-time avoidance system that can sense static and moving objects to prevent vehicle, cyclist and pedestrian collisions, to learning driving policies for an optimal and safe driving experience. We apply deep learning to various challenges of driving including: semantic scene understanding, cross-modal transfer learning, driver awareness understanding, and driver behavior policy.

Now we want to open some of our deep driving challenges to the outside world and invite aspiring researchers to test their chops and potentially even join our team. Today, we are excited to announce our first Nexar Challenge (Using Deep Learning for Traffic Light Recognition), where you can compete to win prizes (1st place $5,000, 2nd place $2,000, 3rd place iPhone 7) and join our mission to make the roads safer.

In the traffic-light challenge, you will take the first step towards a collision-free world by applying deep learning techniques to recognize traffic lights in a car’s driving direction while taking into account the network computational budget.

We can’t wait to see what our challenge participants will come up with, and we hope you will have as much fun as we’ve had while working on the deep-driving challenges here at Nexar.

Let’s Go!
Click Here For More Details