RAR16pub: Recurrent Autoregressive Networks for Online Multi-Object Tracking

MOT16-01


Benchmark:

MOT16 |

Short name:

RAR16pub

Detector:

Public

Description:

Recurrent autoregressive networks tracking framework with public detections

Reference:

K. Fang, Y. Xiang, X. Li, S. Savarese. Recurrent Autoregressive Networks for Online Multi-Object Tracking. In The IEEE Winter Conference on Applications of Computer Vision (WACV), 2018.

Last submitted:

March 24, 2017 (3 years ago)

Published:

July 23, 2018 at 09:33:19 CET

Submissions:

1

Project page / code:

Open source:

No

Hardware:

Titan X, 1.5GHZ, 1Core

Runtime:

0.9 Hz

Benchmark performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT1645.948.874.8100 (13.2)318 (41.9)6,87191,17350.093.01.2648 (13.0)1,992 (39.8)

Detailed performance:

Sequence MOTA IDF1 MOTP MT ML FP FN Recall Precision FAF ID Sw. Frag
MOT16-0126.333.673.0213414,66327.197.70.11249
MOT16-0354.652.874.934223,49243,67958.294.62.33011,072
MOT16-0645.253.972.8311074145,82149.593.20.386205
MOT16-0739.945.973.66196599,07444.491.71.373188
MOT16-0829.835.177.862497110,68836.186.21.689166
MOT16-1239.048.375.312406344,38047.286.10.743131
MOT16-1426.639.174.599366012,86830.489.50.944181

Raw data:


RAR16pub