Mix and match networks: cross-modal alignment for zero-pair image-to-image translation
This addresses a challenge in computer vision for tasks like multimodal translation where training data is incomplete, though it appears incremental as it builds on encoder-decoder alignment methods.
The paper tackles the problem of inferring unseen cross-modal image-to-image translations, such as between depth and semantic segmentation without paired data, by proposing mix and match networks that align encoders and decoders to enable cascading for unseen pairs, showing effectiveness and scalability compared to pairwise approaches.
This paper addresses the problem of inferring unseen cross-modal image-to-image translations between multiple modalities. We assume that only some of the pairwise translations have been seen (i.e. trained) and infer the remaining unseen translations (where training pairs are not available). We propose mix and match networks, an approach where multiple encoders and decoders are aligned in such a way that the desired translation can be obtained by simply cascading the source encoder and the target decoder, even when they have not interacted during the training stage (i.e. unseen). The main challenge lies in the alignment of the latent representations at the bottlenecks of encoder-decoder pairs. We propose an architecture with several tools to encourage alignment, including autoencoders and robust side information and latent consistency losses. We show the benefits of our approach in terms of effectiveness and scalability compared with other pairwise image-to-image translation approaches. We also propose zero-pair cross-modal image translation, a challenging setting where the objective is inferring semantic segmentation from depth (and vice-versa) without explicit segmentation-depth pairs, and only from two (disjoint) segmentation-RGB and depth-RGB training sets. We observe that a certain part of the shared information between unseen modalities might not be reachable, so we further propose a variant that leverages pseudo-pairs which allows us to exploit this shared information between the unseen modalities.