Modeling Images using Transformed Indian Buffet Processes
This work addresses the challenge of modeling unsegmented real images with multiple objects at different locations for computer vision applications, representing an incremental improvement over existing methods.
The authors tackled the problem of modeling real images with transformation-invariant latent features by extending the transformed Indian buffet process (tIBP) to handle real images and developing an efficient inference method, achieving effective image reconstruction in natural images.
Latent feature models are attractive for image modeling, since images generally contain multiple objects. However, many latent feature models ignore that objects can appear at different locations or require pre-segmentation of images. While the transformed Indian buffet process (tIBP) provides a method for modeling transformation-invariant features in unsegmented binary images, its current form is inappropriate for real images because of its computational cost and modeling assumptions. We combine the tIBP with likelihoods appropriate for real images and develop an efficient inference, using the cross-correlation between images and features, that is theoretically and empirically faster than existing inference techniques. Our method discovers reasonable components and achieve effective image reconstruction in natural images.