artitw / interaction-tracker

disease surveillance inspired by robotics/autonomous driving techniques to track the spread of coronavirus

interaction-tracker

Introduction

Motivation for this project: disease surveillance inspired by robotics/autonomous driving techniques to track the spread of coronavirus.

Goals

Hackathon requirements

Related works

Trajectron++

• Formal input and output of network?
• Input to scene representation graph builder?
• How do they visualize?
• What is the reference frame of each agent? Are these then projected onto a ground plane for visualization purposes?

Technical components/ tasks

1. Train Trajectron on NuScenes (Divya)
2. Write inference code - pytorch (Ching Lam)
3. Algorithm to consume output of Trajectron++ network and construct scene representation (graph) (Art)
4. Visualization (if time permits)

Plan of action

Division of labor

Timeline

Deadline: March 16th 5pm PST

Github/links

References

• https://arxiv.org/pdf/2001.03093.pdf Trajectron++