IEEE Smart World NVIDIA AI City Challenge

August 5, 2017, San Francisco, USA

Problem Statement

There will be 1 billion cameras by 2020. Transportation is one of the largest segments that can benefit from actionable insights derived from data captured by these cameras. Between traffic, signaling systems, transportation systems, infrastructure, and transit, the opportunity for insights from these cameras to make transportation systems safer and smarter is immense. Unfortunately, there are several reasons why these potential benefits have not yet materialized for this vertical. Poor data quality, the lack of labels for the data, and the lack of high quality models that can convert the data into actionable insights are some of the biggest impediments to unlocking the value of the data. There is also need for platforms that allow for appropriate analysis from edge to cloud, which will accelerate the development and deployment of these models.

The AI City Challenge

The AI City Challenge by IEEE Smart World and NVIDIA will aim to bring together the following:

1. A traffic camera video dataset of reasonable quality.
2. A collaborative annotation effort among participating teams to label the dataset.
3. Powerful GPU computing infrastructure, from cloud to edge, to enable the training and inferencing of models for real-time detection and classification of objects.
4. A panel of subject matter experts to advise on AI City analytics and help with problems related to safety and congestion.

The Challenge will comprise of two tracks. Teams submitting proposals for participation in either challenge must participate in a collaborative effort to label the data sets.

1. Track 1 - AI City Foundation Track - Teams will use the training set data and labels to develop algorithms and models for basic machine learning tasks, such as object detection, and classification.
2. Track 2 - AI City Applications Track - Teams submitting proposals for participation in this challenge will leverage available datasets to propose and develop AI City applications geared towards solving salient problems related to safety and/or congestion in an urban environment. Teams are encouraged to consider incorporating edge devices if they enhance their solutions.

The challenge will culminate in a one-day workshop at the IEEE Smart World Conference in East Bay Silicon Valley, where participating teams will demonstrate the performance of their systems.

Important Dates

June 1, 2017: Call for Participation
June 19, 2017: Collaborative Annotation Begins
July 5 2017: Collaborative Annotation Ends
July 9 2017: Teams notified of acceptance to Tracks 1 and 2
July 10 2017: Tracks 1 and 2 Begin. Teams receive access to DGX servers, and labeled data
July 31, 2017: Track 1 and Track 2 Submissions and Results Due
August 4, 2017: Demonstrations on Jetson TX2s Due

IEEE Smart World NVIDIA AI City Challenge

Register Online »


logo logo logo logo logo logo

For any questions, please email