README.md

C++

The source code of the project. Code is located in src/.

Compile

Compile everything:

$ make compile [use_caffe=true] [force_rebuild=true]

Remove all created files:

$ make clean

Check the value of a makefile variable:

$ make print-<variable>

NOTE: Calling make <target> without passing use_caffe=true will make the preprocessor remove all code that uses Caffe. This is to enable compilation without Caffe installed. However, this can lead to problems when e.g. make detect use_caffe=true is called after make track, since the files that use Caffe will not be rebuilt. To prevent this, either pass use_caffe=true every time you compile, or call force_rebuild=true to force all .o files to be rebuilt.

Demo

Two example usages are provided. Note that the following targets will also compile the code, and so the same variables (use_caffe and force_rebuild) are used.
Three example usages are provided:

1. Detect - Detect objects in a provided sequence of images. Objects are detected by a CNN (Caffe).

$ make detect
$ ./detectApp.out [-s sequencesFile] [-m modelConfigFile]

2. Track - Track objects from provided detections. The detections should be on the format used in MOTChallenge.

$ make track
$ ./trackApp.out [-s sequencesFile] [-m modelType] \
                    [-d detectionFormat] [-i frameInterval]

3. Detect and Track - Track objects detected in a provided sequence of images. Objects are detected by a CNN (Caffe).

$ make detectAndTrack
$ ./trackApp.out [-s sequencesFile] [-m modelConfigFile]

Integrate in other projects

• To detect objects from images, create an instance of SSDDetector. Requires Caffe. Use function detector.detect(<image>).
• To track objects from pre-existing detections, create an instance of PAOT. No code in detector/ is needed. Use function tracker.track(<detections>).
• To detect and track objects from images, create an instance of ImageTracker. Requires Caffe. Use function tracker.detectAndTrack(<image>).