In the Shadows, Shape Priors Shine: Using Occlusion to Improve Multi-Region Segmentation
This addresses segmentation challenges in computer vision for applications like image analysis, but it is incremental as it builds on existing shape prior methods.
The paper tackles the problem of multi-region segmentation in 2D images with occluding objects by using shape priors and occlusion handling, showing significant improvement over a representative algorithm on preprocessed natural and synthetic images, with good accuracy in recovering ground truth on synthetic images.
We present a new algorithm for multi-region segmentation of 2D images with objects that may partially occlude each other. Our algorithm is based on the observation hat human performance on this task is based both on prior knowledge about plausible shapes and taking into account the presence of occluding objects whose shape is already known - once an occluded region is identified, the shape prior can be used to guess the shape of the missing part. We capture the former aspect using a deep learning model of shape; for the latter, we simultaneously minimize the energy of all regions and consider only unoccluded pixels for data agreement. Existing algorithms incorporating object shape priors consider every object separately in turn and can't distinguish genuine deviation from the expected shape from parts missing due to occlusion. We show that our method significantly improves on the performance of a representative algorithm, as evaluated on both preprocessed natural and synthetic images. Furthermore, on the synthetic images, we recover the ground truth segmentation with good accuracy.