Global and Local Features through Gaussian Mixture Models on Image Semantic Segmentation
This work addresses boundary accuracy issues in semantic segmentation for computer vision applications, presenting an incremental improvement.
The paper tackles the problem of spatial precision loss at object boundaries in semantic segmentation by proposing a model that combines global and local feature representations using Gaussian Mixture Models, showing improvements on Cityscapes and Synthia datasets.
The semantic segmentation task aims at dense classification at the pixel-wise level. Deep models exhibited progress in tackling this task. However, one remaining problem with these approaches is the loss of spatial precision, often produced at the segmented objects' boundaries. Our proposed model addresses this problem by providing an internal structure for the feature representations while extracting a global representation that supports the former. To fit the internal structure, during training, we predict a Gaussian Mixture Model from the data, which, merged with the skip connections and the decoding stage, helps avoid wrong inductive biases. Furthermore, our results show that we can improve semantic segmentation by providing both learning representations (global and local) with a clustering behavior and combining them. Finally, we present results demonstrating our advances in Cityscapes and Synthia datasets.