EMNT: EMNT with RTU ++

MOTS20-01


Benchmark:

MOTS |

Short name:

EMNT

Description:

S. Wang, H. Sheng, D. Yang, Y. Zhang, Y. Wu and S. Wang, "Extendable Multiple Nodes Recurrent Tracking Framework with RTU++," in IEEE Transactions on Image Processing, 2022, doi: 10.1109/TIP.2022.3192706.

Reference:

S. Wang, H. Sheng, D. Yang, Y. Zhang, Y. Wu and S. Wang, "Extendable Multiple Nodes Recurrent Tracking Framework with RTU++," in IEEE Transactions on Image Processing, 2022, doi: 10.1109/TIP.2022.3192706.

Last submitted:

May 19, 2022 (6 months ago)

Published:

June 06, 2022 at 10:26:12 CET

Submissions:

2

Open source:

No

Hardware:

R9-5900X RTX3090

Runtime:

11.7 Hz

Benchmark performance:

Sequence sMOTSA IDF1 MOTSA MOTSP MODSA MT ML TP FP FN Rcll Prcn ID Sw. Frag
MOTS70.077.083.784.184.5234 (71.3)13 (4.0)27,9436664,32686.697.7261 (301.4)618 (713.7)

Detailed performance:

Sequence sMOTSA IDF1 MOTSA MOTSP MODSA MT ML TP FP FN Rcll Prcn ID Sw. Frag
MOTS20-0169.585.184.582.684.7802,6835142386.498.1755
MOTS20-0674.972.088.585.289.814759,01319780191.897.9130239
MOTS20-0765.476.380.482.081.135410,6922532,18683.097.787229
MOTS20-1271.982.382.787.483.34445,55516591685.897.13795

Raw data: