The color and distribution of illuminants can significantly alter the appearance of a scene. The goal of color constancy (CC) is to remove the color bias introduced by the illuminants. Most existing CC algorithms assume a uniformly illuminated scene. However, more often than not, this assumption is an insufficient approximation of realworld illumination conditions (multiple light sources, shadows, interreflections, etc.). Thus, illumination should be locally determined, taking under consideration that multiple illuminants may be present. In this paper we investigate the suitability of adapting 5 state-of-the-art color constancy methods so that they can be used for local illuminant estimation. Given an arbitrary image, we segment it into superpixels of approximately similar color. Each of the methods is applied independently on every superpixel. For improved accuracy, these independent estimates are combined into a single illuminant-color value per superpixel. We evaluated different fusion methodologies. Our experiments indicate that the best performance is obtained by fusion strategies that combine the outputs of the estimators using regression.

No abstract

This Working Paper is issued under the auspices of the Institute for Monetary and Financial Stability (IMFS). Any opinions expressed here are those of the author(s) and not those of the IMFS. Research disseminated by the IMFS may include views on policy, but the IMFS itself takes no institutional policy positions. The Institute for Monetary and Financial Stability aims at raising public awareness of the importance of monetary and financial stability. Its main objective is the implementation of the "Project Monetary and Financial Stability" that is supported by the Foundation of Monetary and Financial Stability. The foundation was established on January 1, 2002 by federal law. Its endowment funds come from the sale of 1 DM gold coins in 2001 that were issued at the occasion of the euro cash introduction in memory of the D-Mark. The IMFS Working Papers often represent preliminary or incomplete work, circulated to encourage discussion and comment. Citation and use of such a paper should take account of its provisional character.

No abstract

This paper analyses the bail-in tool under the BRRD and predicts that it will not reach its policy objective. To make this argument, this paper first describes the policy rationale that calls for mandatory private sector involvement (PSI). From this analysis the key features for an effective bail-in tool can be derived. These insights serve as the background to make the case that the European resolution framework is likely ineffective in establishing adequate market discipline through risk-reflecting prices for bank capital. The main reason for this lies in the avoidable embeddedness of the BRRD's bail-in tool in the much broader resolution process which entails ample discretion of the authorities also in forcing private sector involvement. Finally, this paper synthesized the prior analysis by putting forward an alternative regulatory approach that seeks to disentangle private sector involvement as a precondition for effective bank-resolution as much as possible from the resolution process as such.