A Local Discriminative Model for Background Subtraction

Conventional background subtraction techniques that update a background model online have difficulties with correctly segmenting foreground objects if sudden brightness changes occur. Other methods that learn a global scene model offline suffer from projection errors. To overcome these problems, we present a different approach that is localand discriminative, i.e. for each pixel a classifier is trained to decide whether the pixel belongs to the background or foreground. Such a model requires significantly less tuning effort and shows a better robustness, as we will demonstrate in quantitative experiments on self-created and standard benchmarks. Finally, segmentation is improved significantly by integrating the probabilistic evidence provided by the local classifiers with a graph cut segmentation algorithm.