Tracking By Detection

TUD-Crossing PETS09-S2L2 ETH-Jelmoli ETH-Linthescher ETH-Crossing AVG-TownCentre ADL-Rundle-1 ADL-Rundle-3 KITTI-16 KITTI-19

Short name:

TBD

Benchmark:

Description:

This tracker operates in three stages: First, objects are detected in each frame independently using the DPM object detector by Ross Girshick and Pedro Felzenszwalb. Second, all detections with a positive score are associated to detections in the next frame using appearance and the bounding box overlap. We predict objects to the next frame using a Kalman filter and associate them globally via the Hungarian method for bipartite matching. To gap occlusions and missed detections, we also associate tracklets with each other in a first stage. Similarly to the second stage the Hungarian algorithm is employed but this time based on a occlusion sensitive appearance model and regression of the bounding boxes in one tracklet from the bounding boxes in the other tracklet. The algorithm outputs all associated tracklets with a lifetime longer than three frames.

The reported running time is dominated by the object detection stage.

Default parameters.

Hardware:

2.6 GHz, 16 Cores

Detector:

Public

Processing:

Batch

Last submitted:

September 16, 2014 (3 years ago)

Published:

November 01, 2014 at 03:09:33 CET

Submissions:

1

Open source:

Yes

Project page / code:

Reference:

A. Geiger, M. Lauer, C. Wojek, C. Stiller, R. Urtasun. 3D Traffic Scene Understanding from Movable Platforms. In Pattern Analysis and Machine Intelligence (PAMI), 2014.

Benchmark performance:

MOTAMOTPFAFMTMLFPFNID Sw.FragSpecificationsDetector
15.970.92.66.4 % 47.9 % 14,94334,7771,9391,9632.6 GHz, 16 CoresPublic
IDF1ID PrecisionID Recall
0.00.00.0

Detailed performance:

Sequence MOTA IDF1 MOTP FAF GT MT ML FP FN ID Sw Frag
TUD-Crossing63.20.073.30.61346.2 % 15.4 % 1292472928
PETS09-S2L235.50.069.23.1427.1 % 14.3 % 1,3404,355523480
ETH-Jelmoli29.70.073.21.24517.8 % 26.7 % 5481,1478991
ETH-Linthescher17.00.074.10.11972.0 % 74.6 % 1247,165125131
ETH-Crossing29.40.074.90.12611.5 % 61.5 % 3267425
AVG-TownCentre10.50.069.72.32262.2 % 54.4 % 1,0205,200176215
ADL-Rundle-1-6.90.070.711.03225.0 % 12.5 % 5,4944,193257281
ADL-Rundle-311.90.072.35.7449.1 % 20.5 % 3,5365,157261230
KITTI-1633.10.071.91.3170.0 % 11.8 % 2718056271
KITTI-1910.90.066.61.3624.8 % 30.6 % 1,3473,102313300
Venice-113.70.071.32.41711.8 % 29.4 % 1,1022,732102131

Raw data:


TBD