Performance Analysis of a Foreground Segmentation Neural Network Model
This work addresses foreground segmentation for applications like fraud and anomaly detection, but it is incremental as it builds on an existing method.
The authors conducted an ablation study of FgSegNet_v2 to analyze its stages and proposed a variation that surpasses state-of-the-art results, achieving overall improvement in the CDNet2014 dataset, particularly in the LowFrameRate subset, and comparable results on SBI2015 and CityScapes datasets under varying conditions.
In recent years the interest in segmentation has been growing, being used in a wide range of applications such as fraud detection, anomaly detection in public health and intrusion detection. We present an ablation study of FgSegNet_v2, analysing its three stages: (i) Encoder, (ii) Feature Pooling Module and (iii) Decoder. The result of this study is a proposal of a variation of the aforementioned method that surpasses state of the art results. Three datasets are used for testing: CDNet2014, SBI2015 and CityScapes. In CDNet2014 we got an overall improvement compared to the state of the art, mainly in the LowFrameRate subset. The presented approach is promising as it produces comparable results with the state of the art (SBI2015 and Cityscapes datasets) in very different conditions, such as different lighting conditions.