LP_SSVM: Linear Programming Learned With Structured SVM

TUD-Crossing


Rendering of new sequences is currently deactivated due to heavy load.

Rendering of new sequences is currently deactivated due to heavy load.

Rendering of new sequences is currently deactivated due to heavy load.

Benchmark:

MOT15 |

Short name:

LP_SSVM

Detector:

Public

Description:

Min-cost network flow model with pairwise interactions; network cost is learned using structured SVM.

Reference:

S. Wang, C. Fowlkes. Learning Optimal Parameters for Multi-target Tracking with Contextual Interactions. In International Journal of Computer Vision, 2016.

Last submitted:

April 20, 2015 (7 years ago)

Published:

April 21, 2015 at 04:22:07 CET

Submissions:

1

Project page / code:

n/a

Open source:

No

Hardware:

2.4GHz, 1 Core

Runtime:

41.3 Hz

Benchmark performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
MOT1525.234.026.342 (5.8)382 (53.0)8,36936,93239.974.526.326.729.066.230.456.975.61.4646 (16.2)849 (21.3)

Detailed performance:

Sequence MOTA IDF1 HOTA MT ML FP FN Rcll Prcn AssA DetA AssRe AssPr DetRe DetPr LocA FAF ID Sw. Frag
ADL-Rundle-114.038.330.2773,5074,42352.558.232.228.935.069.340.044.475.97.06992
ADL-Rundle-328.036.928.54101,8555,38847.072.027.130.529.069.036.255.476.23.08183
AVG-TownCentre14.724.319.761394595,51522.878.123.416.928.753.617.961.073.71.0123141
ETH-Crossing24.937.526.01171074126.196.334.319.935.380.620.375.079.20.022
ETH-Jelmoli39.552.435.58162241,29349.084.736.734.641.069.838.165.977.90.51729
ETH-Linthescher15.623.521.05157417,48316.297.235.712.437.976.012.575.379.40.01115
KITTI-1639.241.727.3029092445.789.623.332.326.957.934.768.077.40.42029
KITTI-1928.239.627.14188102,95544.774.725.329.629.758.733.355.671.90.870162
PETS09-S2L241.527.921.1376294,80350.288.512.935.213.959.837.666.274.61.4212249
TUD-Crossing60.055.038.3424837566.093.831.846.238.657.349.870.877.70.21820
Venice-117.831.925.0076963,03233.668.728.022.528.976.926.153.476.61.52327

Raw data: