PatchVAE: Learning Local Latent Codes for Recognition
This work addresses the gap between unsupervised and supervised representation learning for recognition, offering an incremental improvement for computer vision applications.
The paper tackles the problem of unsupervised representation learning for recognition by proposing PatchVAE, a method that learns local latent codes at the patch level to capture repeating patterns, resulting in significantly better performance on recognition tasks compared to vanilla VAEs.
Unsupervised representation learning holds the promise of exploiting large amounts of unlabeled data to learn general representations. A promising technique for unsupervised learning is the framework of Variational Auto-encoders (VAEs). However, unsupervised representations learned by VAEs are significantly outperformed by those learned by supervised learning for recognition. Our hypothesis is that to learn useful representations for recognition the model needs to be encouraged to learn about repeating and consistent patterns in data. Drawing inspiration from the mid-level representation discovery work, we propose PatchVAE, that reasons about images at patch level. Our key contribution is a bottleneck formulation that encourages mid-level style representations in the VAE framework. Our experiments demonstrate that representations learned by our method perform much better on the recognition tasks compared to those learned by vanilla VAEs.