Layered Graphical Models dataset
We introduce a dataset for tracking pedestrians in the vicinity of
relocatable occluders, e.g., parked vehicles.
Related publications
[1] Vitaly Ablavsky and Stan Sclaroff, Layered Graphical Models for
Tracking partiall-Occluded Moving Objects, IEEE Trans. on Pattern
Analysis and Machine Intelligence (PAMI), published online March 2011
(pdf)
[2] Vitaly Ablavsky, Ashwin Thangali, and Stan Sclaroff, Layered Graphical Models for
Tracking partiall-Occluded Objects, In Proc., IEEE Conf. Computer
Vision and Pattern Recognition (CVPR), 2008 (pdf)
Data
Video sequences in the dataset are described in [1], and the
naming convention follows Table 4 of Sec. 4.2.
Preview-quality videos (Xvid MPEG-4) may be found here:
| seq. "a" |
seq. "b" |
seq. "c" |
 |
 |
 |
| seq. "d" |
seq. "e" (from PETS 2001) |
seq. "f" |
 |
 |
 |
Source videos may be found here
(uncompressed avi)
Example results from [1]
These videos demonstrate tracking pedestrians in the vicinity
of relocatable occluders (i.e., vehicles) at rest. Pedestrians
farther away from the parked vehicles may be tracked using
standard approaches.
| seq. "a" |
seq. "e" (Frames 1 and 950 show the corresponding state space;
cf. Fig 7 of [1].) |
 |
 |