Visual Scene Representations: Contrast, Scaling and Occlusion
This work provides a theoretical foundation for visual scene representations, connecting to classical theories, but it is incremental as it builds on existing descriptors without broad empirical validation.
The paper tackled the problem of deriving minimal sufficient statistics for visual data under scaling and occlusion, showing that under restrictive assumptions, these representations relate to existing computer vision features, highlighting tacit conditions and suggesting improvements.
We study the structure of representations, defined as approximations of minimal sufficient statistics that are maximal invariants to nuisance factors, for visual data subject to scaling and occlusion of line-of-sight. We derive analytical expressions for such representations and show that, under certain restrictive assumptions, they are related to features commonly in use in the computer vision community. This link highlights the condition tacitly assumed by these descriptors, and also suggests ways to improve and generalize them. This new interpretation draws connections to the classical theories of sampling, hypothesis testing and group invariance.