We are introducing AI in Dhaka Traffic System

For the first time in Bangladesh, an international AI-based Dhaka Traffic Detection Challenge funded by Elsevier would be co-organized during STI 2020

Register Talk with DhakaAI

The capital city of Dhaka has only 7% traffic roads (compared to 25% urban standard) in presence of approximately 8 million commuters a day within 306 sq km area. The scenario of Dhaka traffic is unique which poses complex new challenges in terms of automated traffic detection. To solve the problem using advances in AI-based technology and ICT solutions, we are calling for solutions to automatic Dhaka traffic detection problems on optical images. This new AI-Based Dhaka Traffic Detection Challenge aims at assessing the ability of state-of-the-art methods to detect and recognize traffic vehicles. This solution is encountered in modern cities where multiple cultures live and communicate together, where users see various scripts and languages in a way that prevents using much a priori knowledge. Also, at the same time, academics and researchers from the region who are experts in AI or interested in exploring possibilities could be brought to a networking community through this campaign. Working together on a common problem statement can create the right synergies needed to build an AI-based community in South-East Asia.

What is

this challenge?

The competition will happen online and within 2 rounds. In the first round, a training dataset would be provided with which the participants need to train a generalized object detection model to locate traffic vehicles and identify them on the 1st test dataset and generate a submission file following the prescribed format. Based only on the detection accuracy, the top 30% team will move to the 2nd round where they will be evaluated based on the detection accuracy on the 2nd test dataset, poster, presentation, and codes.

Dataset

description

The dataset is composed of vehicle images, where an image contains a vehicle of one or more of 21 different classes of vehicle. This makes the dataset useful for multiple vehicle detection and recognition. The considered vehicle classes are: ambulance, auto-rickshaw, bicycle, bus, car, garbage van, human hauler, minibus, minivan, motorbike, Pickup, army vehicle, police car, rickshaw, scooter, Suv, taxi, three-wheelers (CNG), truck, van, wheelbarrow.

Dataset also available in Kaggle

Our Sponsors

Registered Teams: 164