SSL_MOT: Self-Supervised Learning Based Multi-Object Tracking

MOT20-04


Benchmark:

MOT16 | MOT17 | MOT20 |

Short name:

SSL_MOT

Detector:

Public

Description:

Reference:

Last submitted:

September 06, 2021 (1 month ago)

Published:

August 24, 2021 at 10:25:38 CET

Submissions:

2

Project page / code:

n/a

Open source:

No

Hardware:

RTX2080ti

Runtime:

1.6 Hz

Benchmark performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT2048.137.530.9421 (33.9)240 (19.3)70,716186,77963.982.421.844.334.734.751.466.280.015.811,247 (0.0)9,024 (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
MOT20-0460.641.734.23164840,34159,94678.184.121.854.236.434.863.668.581.119.47,7984,849
MOT20-0633.828.923.9489216,55469,30147.879.317.532.927.828.237.562.277.816.42,0922,374
MOT20-0761.452.341.937131,75310,74967.592.736.049.445.552.753.873.880.73.0283418
MOT20-0822.727.522.7208712,06846,78339.671.819.826.527.833.330.855.876.415.01,0741,383

Raw data:


SSL_MOT