RETracker: Random Ensemble Tracker

MOT20-04


Benchmark:

MOT20 |

Short name:

RETracker

Detector:

Public

Description:

Reference:

Y. Kawanishi. Label-Based Multiple Object Ensemble Tracking with Randomized Frame Dropping. In Proceedings of the 26th International Conference on Pattern Recognition, 2022.

Last submitted:

January 05, 2022 (8 months ago)

Published:

September 19, 2022 at 12:37:16 CET

Submissions:

2

Project page / code:

n/a

Open source:

No

Hardware:

3.00 GHz 18 core

Runtime:

0.0 Hz

Benchmark performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT2062.453.045.3605 (48.7)192 (15.5)43,503147,45171.589.538.653.453.352.359.374.282.19.73,804 (0.0)9,245 (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
MOT20-0475.961.252.54074520,45744,44483.891.843.264.061.055.370.377.082.59.81,0173,233
MOT20-0646.640.633.8867312,21557,16956.986.127.841.337.244.146.269.981.212.11,5503,379
MOT20-0776.061.252.07861,9715,73882.793.343.063.154.456.869.378.283.93.4224421
MOT20-0835.535.830.434688,86040,10048.280.827.733.735.545.638.464.380.011.01,0132,212

Raw data: