Visual Tracking using Global and Local visual information

Introduction

We addresses the problem of tracking objects which undergo rapid and significant appearance changes. We propose a novel coupled-layer visual model that combines the target's global and local appearance.

The local layer in this model is a set of local patches that geometrically constrain the changes in the target's appearance. This layer probabilistically adapts to the target's geometric deformation, while its structure is updated by removing and adding the local patches. The addition of the patches is constrained by the global layer that probabilistically models target's global visual properties such as color, shape and apparent local motion. The global visual properties are updated during tracking using the stable patches from the local layer. By this coupled constraint paradigm between the adaptation of the global and the local layer, we achieve a more robust tracking through significant appearance changes.

Our experimental results on challenging sequences confirm that our tracker outperforms the related state-of-the-art trackers by having smaller failure rate as well as better accuracy.

Publications

Publications for the topic of Local-Global Tracking

Source code

Source code for the ICCV 2011 paper An adaptive coupled-layer visual model for robust visual tracking - Updated on 12/11/2012
It is recommended that you use the TPAMI code below instead as it is more stable and still (to some degree) actively maintained.
Source code for the TPAMI journal article Robust Visual Tracking using an Adaptive Coupled-layer Visual Model - first version 18/03/2013.
A complete Matlab re-implementation of the LGT tracker with improved parameters is now part of this Github repository.
A C++ implementation is available as a part of the the Legit visual tracking library available on Github.

Tracking sequences and annotations

Here we attach our own sequences used in our experiments:

Stationary camera, toy dinosaur moved around the scene. From the ICCV2011 paper An adaptive coupled-layer visual model for robust visual tracking.
Articulated object, fast motion and deformations, moving camera. From the ICCV2011 paper An adaptive coupled-layer visual model for robust visual tracking.
Articulated, deformable object, stationary camera. From the ICCV2011 paper An adaptive coupled-layer visual model for robust visual tracking.
Articulated, deformable object, stationary camera. From the ICCV2011 paper An adaptive coupled-layer visual model for robust visual tracking.
Hollow shape, uniform color. From the ICCV2011 paper An adaptive coupled-layer visual model for robust visual tracking.
Stationary camera, out of plane rotations, blur. From the TPAMI paper Robust Visual Tracking using an Adaptive Coupled-layer Visual Model.
Articulated object, camera motion, occlusion. From the TPAMI paper Robust Visual Tracking using an Adaptive Coupled-layer Visual Model.

Other