A Differential Invariant for Zooming

TL;DR

This paper introduces a local differential invariant that remains unchanged under zooming and brightness adjustments, utilizing differential Gaussian operators for robust feature detection during camera zoom-out scenarios.

Contribution

The paper proposes a novel differential invariant based on Gaussian operators that is invariant to zooming and brightness changes, enhancing feature stability in visual analysis.

Findings

Invariant effectively detects features during zoom-out simulations

Differential Gaussian operators up to third order improve robustness

Invariant maintains stability under linear brightness changes

Abstract

This paper presents an invariant under scaling and linear brightness change. The invariant is based on differentials and therefore is a local feature. Rotationally invariant 2-d differential Gaussian operators up to third order are proposed for the implementation of the invariant. The performance is analyzed by simulating a camera zoom-out.