Shape complexity estimation using VAE
This work addresses shape complexity estimation, which is an incremental contribution to computer vision and graphics.
The paper tackled the problem of estimating shape complexity by comparing methods and introducing a new approach using Variational Autoencoders (VAEs) with different latent vector sizes, demonstrating that it captures some aspects of shape complexity.
In this paper, we compare methods for estimating the complexity of two-dimensional shapes and introduce a method that exploits reconstruction loss of Variational Autoencoders with different sizes of latent vectors. Although complexity of a shape is not a well defined attribute, different aspects of it can be estimated. We demonstrate that our methods captures some aspects of shape complexity. Code and training details will be publicly available.