Stable multi-target tracking in real-time surveillance video
Citations
1,604 citations
Cites background from "Stable multi-target tracking in rea..."
...This category comprise of three videos from [1] with a fast-moving motorbike in the desert, a low contrast recording of a car on the highway, and a car-chase; three videos from the 3DPeS dataset [28] with varying illumination conditions and with complex object motion; one video from the dataset in [107] with surveillance of multiple people; and three complex videos from YouTube....
[...]
667 citations
Cites methods from "Stable multi-target tracking in rea..."
...9 yes static high cloudy [10] ADL-Rundle-1 30 1920x1080 500 (00:17) 32 9306 18....
[...]
...For the 4 sequences filmed using a static camera, AVG-TownCentre, PETS09-S2L1, PETS09-S2L2 and TUDStadtmitte, the calibration files from the sources [5], [10], [20] are used to compute a 2D homography between the image plane and the ground plane....
[...]
616 citations
547 citations
411 citations
References
31,952 citations
"Stable multi-target tracking in rea..." refers methods in this paper
...We make motion estimates by following corner features with pyramidal Kanade-Lucas-Tomasi (KLT) tracking [15, 8]....
[...]
12,944 citations
2,286 citations
"Stable multi-target tracking in rea..." refers background or methods in this paper
...Multiple Object Tracking Accuracy (MOTA) is a combined measure which takes into account false positives, false negatives and identity switches (see [5] for details)....
[...]
...The Multiple Object Tracking Precision (MOTP) measures the precision with which objects are located using the intersection of the estimated region with the ground truth region....
[...]
...A comprehensive evaluation of the realtime tracker on two different datasets is performed using the standard CLEAR MOT [5] evaluation criteria....
[...]
...Both resulted in a significant reduction in the MOTA and either the precision or recall....
[...]
637 citations
"Stable multi-target tracking in rea..." refers background in this paper
...The first group covers feed-forward systems which use only current and past observations to estimate the current state [7, 1, 2]. The second group covers data association based methods which also use future information to estimate the current state, allowing ambiguities to be more easily resolved at the cost of increased latency [13, 11, 12 , 3, 20]....
[...]