Traffic4cast Symposium 2019

IARAI Traffic4cast Full-Day Symposium, 15 Dec, 9am – 5pm

Time	Topic	Speaker	Affiliation
9:00	Introduction	David Kreil	IARAI
9:05	Optimal Delayed Reward Redistribution	Sepp Hochreiter	IARAI, Johannes Kepler University
9:45	Early Activity Detection and Semantic Understanding in Videos	Leonid Sigal	University of British Columbia
10:15	The Making Of – Compiling the Large-scale Real-world Data for Traffic4cast	Moritz Neun	HERE Technologies
11:00	Traffic Map Movie Prediction Using U-Net with DenseNet	Sungbin Choi		abstract
11:30	Augmenting the Traffic4cast Challenge 2019 with Spatio-temporal Context	Henry Martin	MIE-Lab, ETH Zurich	abstract
12:00	CrevNet: Conditionally Reversible Video Prediction	Wei Yu	University of Toronto	abstract
12:30	Building an Effective Large-Scale Traffic State Prediction System	Qiming Zhang	University of Sydney	abstract
13:00	Recurrent Autoencoder with Skip Connections and Exogenous Variables for Traffic Forecasting	Pedro Herruzo Sánchez	Polytechnic University of Catalonia	abstract
14:30	Spatiotemporal Tile-based Attention-guided LSTMs for Traffic Video Prediction	Tu Nguyen	Daimler Autonomous Services	abstract
15:00	Traffic4Cast – Aligning Short-term Traffic State Forecasting and Video Frame Prediction	David Jonietz	HERE Technologies
15:50	Significance and Relevance of Leaderboard Ranks	Aleksandra Gruca	Silesian University of Technology
16:20	Picking the Right Loss – Choice of an Appropriate Metric	Michael Kopp	IARAI, HERE Technologies
16:50	Outlook to 2020	David Kreil	IARAI
17:00	Closing

Session Chairs: David Kreil (IARAI), Moritz Neun (HERE), Michael Kopp (IARAI & HERE), Shia Shi (HERE).

Traffic4cast Competition Track at NeurIPS 2019, 14 Dec.

Venue: Fairmount Vancouver hotel (5 mins walking from NeurIPS, down Burrard street). The meeting was held in the Salt Spring Island Room on ‘Floor 1’ of the hotel, with expansive windows overlooking the historic Vancouver Art Gallery.