unizard / AwesomeArxiv

89 stars 14 forks source link

[2018.07.02] End-to-end Learning of Multi-sensor 3D Tracking by Detection #200

Open unizard opened 5 years ago

unizard commented 5 years ago

ICRA2018

Company: UBER URL: https://arxiv.org/pdf/1806.11534.pdf Interest: 4 Keyword: RGB-D Tracking, MOTA, DeepLearning

In this paper they propose a novel approach to tracking by detection that can exploit both cameras as well as LIDAR data to produce very accurate 3D trajectories.

unizard commented 5 years ago

Proposed Architecture
Visualization of a set of trajectories produced by the tracker over 15 frames. Trajectories are color coded, such that having the same color means it’s the same object.
Qualitative Results

mjjdick commented 5 years ago

great job！

gebawe commented 3 years ago

@unizard

I like your implementation. Can you please share the code?