GMPHD_Rd20: Occlusion-robust Online Multi-object Visual Tracking using a GM-PHD Filter with a CNN-based Re-identification

MOT20-04


Benchmark:

MOT20 |

Short name:

GMPHD_Rd20

Detector:

Public

Description:

Reference:

N. Baisa. Occlusion-robust online multi-object visual tracking using a GM-PHD filter with CNN-based re-identification. In Journal of Visual Communication and Image Representation, 2021.

Last submitted:

November 22, 2020 (1 year ago)

Published:

November 15, 2020 at 12:02:52 CET

Submissions:

2

Project page / code:

n/a

Open source:

No

Hardware:

3 GHZ, 1 Core

Runtime:

25.2 Hz

Benchmark performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT2044.743.535.6293 (23.6)274 (22.1)42,778236,11654.486.832.039.934.172.044.070.380.29.67,492 (137.8)11,153 (205.1)

Detailed performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT20-0463.650.241.5215667,39788,07367.996.233.052.535.172.355.678.882.53.64,4256,001
MOT20-0623.634.326.9299918,44881,26438.873.627.327.029.069.530.658.175.618.31,6772,886
MOT20-0750.151.239.937191,86414,20457.191.039.041.541.575.745.071.879.93.2438578
MOT20-0811.529.223.4129015,06952,57532.162.325.821.627.764.525.449.373.818.79521,688

Raw data: