OVBT17: Tracking Multiple Persons Based on a Variational Bayesian Model

MOT20-04


Benchmark:

MOT17 | MOT20 |

Short name:

OVBT17

Detector:

Public

Description:

Reference:

Y. Ban, S. Ba, X. Alameda-Pineda, R. Horaud. Tracking multiple persons based on a variational bayesian model. In European Conference on Computer Vision Workshops, 2016.

Last submitted:

August 01, 2022 (4 months ago)

Published:

August 01, 2022 at 16:04:16 CET

Submissions:

2

Project page / code:

n/a

Open source:

No

Hardware:

Intel® Core™ i9-9820X CPU @ 3.30GHz × 20

Runtime:

0.1 Hz

Benchmark performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT2040.037.830.5141 (11.4)374 (30.1)23,368282,94945.390.927.434.431.060.636.773.781.15.24,210 (0.0)10,026 (0.0)

Detailed performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT20-0455.644.936.01041043,069116,03857.798.128.745.633.159.347.580.783.11.52,4985,885
MOT20-0622.627.321.81613110,01291,78930.980.421.622.223.458.724.062.676.69.91,0002,343
MOT20-0744.745.934.9163086517,24147.994.834.136.336.175.538.375.781.31.5203471
MOT20-0812.523.218.751099,42257,88125.367.520.117.821.757.619.953.074.511.75091,327

Raw data: