Object tracking project
Groups with 2 students need to implement only:
- Tracking of objects in a simple sequence
- Management of object identity
- Evaluation of results, e.g., relative to ground truth
Groups with 3 students must, in addition to the above, implement functionality that can deal with either of
- shadows / reflections
- more than one background model to manage spurious motions
- occlusions, e.g., using predition or ground plane modeling
There are several data sets for motion analysis available on the internet, but we recommend that you use the following
In addition to these, there is also a dataset produced in the IVSS project which is available only from our local file system, e.g., if you are in the student computer halls at ISY. These sequences contain signficant amount of shadows, both in terms of the moving objects that cast shadows on the background and features of the background that cast shadows onto the moving objects. The path to the sequences are:
The sequences in this folder are
- harder1, images 1 to 2800. With changing lighting conditions, from overcast to moving scattered clouds. Shadows appearing and disappering. People fetching parked trucks.
- harder2, images 1 to 941. An interesting challenge, don't start with this one!
- Renova_20070907_081432_Cam10_0005, images 1 to 1901. Shadows.
- Renova_20080420_083025_Cam10_0000, images 1 to 1101. Shadows.
- Renova_20080618_083045_Cam10_0002, images 1 to 2101. Almost no shadows, no reflections, interesting occlusion starting at frame 1255. Suitable for something to start with.
If you have set up the paths in Matlab properly, each image can be loaded, e.g., by
>> imread(sprintf('dataRenova_20080420_083025_Cam10_0000_%05d.jpg', frameid));
- John Wood, Statistical Background Models with Shadow Detection for Video Based Tracking, Master Thesis 2007. Linköping University
- Håkan Ardö, Multi-target Tracking Using on-line Viterbi Optimisation and Stochastic Modelling, PhD thesis, 2009, Lund University