Spotlight the Negatives: A Generalized Discriminative Latent Model
This work addresses a specific limitation in visual recognition systems for computer vision researchers, offering an incremental enhancement to existing methods.
The paper tackles the problem of improving discriminative latent variable models for visual recognition by introducing negative latent variables for the background class, resulting in significant improvements on two detection tasks.
Discriminative latent variable models (LVM) are frequently applied to various visual recognition tasks. In these systems the latent (hidden) variables provide a formalism for modeling structured variation of visual features. Conventionally, latent variables are de- fined on the variation of the foreground (positive) class. In this work we augment LVMs to include negative latent variables corresponding to the background class. We formalize the scoring function of such a generalized LVM (GLVM). Then we discuss a framework for learning a model based on the GLVM scoring function. We theoretically showcase how some of the current visual recognition methods can benefit from this generalization. Finally, we experiment on a generalized form of Deformable Part Models with negative latent variables and show significant improvements on two different detection tasks.