Semantics-Guided Representation Learning with Applications to Visual Synthesis
This work addresses the challenge of learning meaningful semantic representations for researchers in computer vision, though it appears incremental as it builds on existing interpolation methods.
The paper tackles the problem of learning latent representations that are both interpretable and interpolatable for visual synthesis, proposing a method that achieves semantic-oriented and visually-smooth interpolation, with experiments on MNIST and CMU Multi-PIE datasets verifying its effectiveness.
Learning interpretable and interpolatable latent representations has been an emerging research direction, allowing researchers to understand and utilize the derived latent space for further applications such as visual synthesis or recognition. While most existing approaches derive an interpolatable latent space and induces smooth transition in image appearance, it is still not clear how to observe desirable representations which would contain semantic information of interest. In this paper, we aim to learn meaningful representations and simultaneously perform semantic-oriented and visually-smooth interpolation. To this end, we propose an angular triplet-neighbor loss (ATNL) that enables learning a latent representation whose distribution matches the semantic information of interest. With the latent space guided by ATNL, we further utilize spherical semantic interpolation for generating semantic warping of images, allowing synthesis of desirable visual data. Experiments on MNIST and CMU Multi-PIE datasets qualitatively and quantitatively verify the effectiveness of our method.