Method

TrackMPNN: A Message Passing Graph Neural Architecture for Multi-Object Tracking [on] [TrackMPNN]
https://github.com/arangesh/TrackMPNN

Submitted on 9 Jan. 2021 00:24 by
Akshay Rangesh (UC San Diego)

Running time:0.05 s
Environment:4 cores @ 3.0 Ghz (Python)

Method Description:
A framework based on dynamic undirected graphs that
represent the data association problem over multiple
timesteps, and a message passing graph neural
network (GNN) that operates on these graphs to
produce the desired likelihood for every association
therein. We only use the 2D detection box location
and score as the descriptor for each object
instance.
Parameters:
N/A
Latex Bibtex:
@article{rangesh2101trackmpnn,
title={TrackMPNN: A Message Passing Graph Neural
Architecture for Multi-Object Tracking},
author={Rangesh, Akshay and Maheshwari, Pranav
and Gebre, Mez and Mhatre, Siddhesh and Ramezani,
Vahid and Trivedi, Mohan M},
journal={arXiv preprint arXiv:2101.04206}
}

Detailed Results

From all 29 test sequences, our benchmark computes the HOTA tracking metrics (HOTA, DetA, AssA, DetRe, DetPr, AssRe, AssPr, LocA) [1] as well as the CLEARMOT, MT/PT/ML, identity switches, and fragmentation [2,3] metrics. The tables below show all of these metrics.


Benchmark HOTA DetA AssA DetRe DetPr AssRe AssPr LocA
CAR 72.30 % 74.69 % 70.63 % 80.02 % 83.11 % 73.58 % 87.14 % 86.14 %
PEDESTRIAN 39.40 % 44.24 % 35.45 % 50.78 % 64.58 % 38.98 % 69.80 % 77.56 %

Benchmark TP FP FN
CAR 31815 2577 1298
PEDESTRIAN 15445 7705 2758

Benchmark MOTA MOTP MODA IDSW sMOTA
CAR 87.33 % 84.49 % 88.73 % 481 72.98 %
PEDESTRIAN 52.10 % 73.42 % 54.80 % 626 34.37 %

Benchmark MT rate PT rate ML rate FRAG
CAR 84.46 % 13.38 % 2.15 % 237
PEDESTRIAN 35.05 % 46.05 % 18.90 % 669

Benchmark # Dets # Tracks
CAR 33113 1236
PEDESTRIAN 18203 870

This table as LaTeX