Debarati Kundu

Bio: Debarati Kundu is an academic researcher from University of Texas at Austin. The author has contributed to research in topics: Image quality & Particle swarm optimization. The author has an hindex of 15, co-authored 23 publications receiving 622 citations. Previous affiliations of Debarati Kundu include Yahoo! & Jadavpur University.

No-Reference Quality Assessment of Tone-Mapped HDR Pictures

Abstract: Being able to automatically predict digital picture quality, as perceived by human observers, has become important in many applications where humans are the ultimate consumers of displayed visual information. Standard dynamic range (SDR) images provide 8 b/color/pixel. High dynamic range (HDR) images, which are usually created from multiple exposures of the same scene, can provide 16 or 32 b/color/pixel, but must be tonemapped to SDR for display on standard monitors. Multi-exposure fusion techniques bypass HDR creation, by fusing the exposure stack directly to SDR format while aiming for aesthetically pleasing luminance and color distributions. Here, we describe a new no-reference image quality assessment (NR IQA) model for HDR pictures that is based on standard measurements of the bandpass and on newly conceived differential natural scene statistics (NSS) of HDR pictures. We derive an algorithm from the model which we call the HDR IMAGE GRADient-based Evaluator. NSS models have previously been used to devise NR IQA models that effectively predict the subjective quality of SDR images, but they perform significantly worse on tonemapped HDR content. Toward ameliorating this we make here the following contributions: 1) we design HDR picture NR IQA models and algorithms using both standard space-domain NSS features as well as novel HDR-specific gradient-based features that significantly elevate prediction performance; 2) we validate the proposed models on a large-scale crowdsourced HDR image database; and 3) we demonstrate that the proposed models also perform well on legacy natural SDR images. The software is available at: http://live.ece.utexas.edu/research/Quality/higradeRelease.zip .

Large-Scale Crowdsourced Study for Tone-Mapped HDR Pictures

Abstract: Measuring digital picture quality, as perceived by human observers, is increasingly important in many applications in which humans are the ultimate consumers of visual information. Standard dynamic range (SDR) images provide 8 b/color/pixel. High dynamic range (HDR) images, usually created from multiple exposures of the same scene, can provide 16 or 32 b/color/pixel, but need to be tonemapped to SDR for display on standard monitors. Multiexposure fusion (MEF) techniques bypass HDR creation by fusing an exposure stack directly to SDR images to achieve aesthetically pleasing luminance and color distributions. Many HDR and MEF databases have a relatively small number of images and human opinion scores, obtained under stringently controlled conditions, thereby limiting realistic viewing. Moreover, many of these databases are intended to compare tone-mapping algorithms, rather than being specialized for developing and comparing image quality assessment models. To overcome these challenges, we conducted a massively crowdsourced online subjective study. The primary contributions described in this paper are: 1) the new ESPL-LIVE HDR Image Database that we created containing diverse images obtained by tone-mapping operators and MEF algorithms, with and without post-processing; 2) a large-scale subjective study that we conducted using a crowdsourced platform to gather more than 300 000 opinion scores on 1811 images from over 5000 unique observers; and 3) a detailed study of the correlation performance of the state-of-the-art no-reference image quality assessment algorithms against human opinion scores of these images. The database is available at http://signal.ece.utexas.edu/%7Edebarati/HDRDatabase.zip .

Inertia-Adaptive Particle Swarm Optimizer for Improved Global Search

Abstract: This paper describes a method for improving the final accuracy and the convergence speed of Particle Swarm Optimization (PSO) by adapting its inertia factor in the velocity updating equation and also by adding a new coefficient to the position updating equation. These modifications do not impose any serious requirements on the basic algorithm in terms of the number of Function Evaluations (FEs). The new algorithm has been shown to be statistically significantly better than four recent variants of PSO on an eight-function test-suite for the following performance matrices: Quality of the final solution, time to find out the solution, frequency of hitting the optima, and scalability.

Multi-Objective Differential Evolution for Automatic Clustering with Application to Micro-Array Data Analysis

Abstract: This paper applies the Differential Evolution (DE) algorithm to the task of automatic fuzzy clustering in a Multi-objective Optimization (MO) framework. It compares the performances of two multi-objective variants of DE over the fuzzy clustering problem, where two conflicting fuzzy validity indices are simultaneously optimized. The resultant Pareto optimal set of solutions from each algorithm consists of a number of non-dominated solutions, from which the user can choose the most promising ones according to the problem specifications. A real-coded representation of the search variables, accommodating variable number of cluster centers, is used for DE. The performances of the multi-objective DE-variants have also been contrasted to that of two most well-known schemes of MO clustering, namely the Non Dominated Sorting Genetic Algorithm (NSGA II) and Multi-Objective Clustering with an unknown number of Clusters K (MOCK). Experimental results using six artificial and four real life datasets of varying range of complexities indicate that DE holds immense promise as a candidate algorithm for devising MO clustering schemes.

Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Abstract: We consider image transformation problems, where an input image is transformed into an output image. Recent methods for such problems typically train feed-forward convolutional neural networks using a per-pixel loss between the output and ground-truth images. Parallel work has shown that high-quality images can be generated by defining and optimizing perceptual loss functions based on high-level features extracted from pretrained networks. We combine the benefits of both approaches, and propose the use of perceptual loss functions for training feed-forward networks for image transformation tasks. We show results on image style transfer, where a feed-forward network is trained to solve the optimization problem proposed by Gatys et al. in real-time. Compared to the optimization-based method, our network gives similar qualitative results but is three orders of magnitude faster. We also experiment with single-image super-resolution, where replacing a per-pixel loss with a perceptual loss gives visually pleasing results.

Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Abstract: We consider image transformation problems, where an input image is transformed into an output image. Recent methods for such problems typically train feed-forward convolutional neural networks using a \emph{per-pixel} loss between the output and ground-truth images. Parallel work has shown that high-quality images can be generated by defining and optimizing \emph{perceptual} loss functions based on high-level features extracted from pretrained networks. We combine the benefits of both approaches, and propose the use of perceptual loss functions for training feed-forward networks for image transformation tasks. We show results on image style transfer, where a feed-forward network is trained to solve the optimization problem proposed by Gatys et al in real-time. Compared to the optimization-based method, our network gives similar qualitative results but is three orders of magnitude faster. We also experiment with single-image super-resolution, where replacing a per-pixel loss with a perceptual loss gives visually pleasing results.

Multiobjective evolutionary algorithms: A survey of the state of the art

Abstract: A multiobjective optimization problem involves several conflicting objectives and has a set of Pareto optimal solutions. By evolving a population of solutions, multiobjective evolutionary algorithms (MOEAs) are able to approximate the Pareto optimal set in a single run. MOEAs have attracted a lot of research effort during the last 20 years, and they are still one of the hottest research areas in the field of evolutionary computation. This paper surveys the development of MOEAs primarily during the last eight years. It covers algorithmic frameworks such as decomposition-based MOEAs (MOEA/Ds), memetic MOEAs, coevolutionary MOEAs, selection and offspring reproduction operators, MOEAs with specific search methods, MOEAs for multimodal problems, constraint handling and MOEAs, computationally expensive multiobjective optimization problems (MOPs), dynamic MOPs, noisy MOPs, combinatorial and discrete MOPs, benchmark problems, performance indicators, and applications. In addition, some future research issues are also presented.

A novel particle swarm optimization algorithm with adaptive inertia weight

Abstract: Particle swarm optimization (PSO) is a stochastic population-based algorithm motivated by intelligent collective behavior of some animals. The most important advantages of the PSO are that PSO is easy to implement and there are few parameters to adjust. The inertia weight (w) is one of PSO's parameters originally proposed by Shi and Eberhart to bring about a balance between the exploration and exploitation characteristics of PSO. Since the introduction of this parameter, there have been a number of proposals of different strategies for determining the value of inertia weight during a course of run. This paper presents the first comprehensive review of the various inertia weight strategies reported in the related literature. These approaches are classified and discussed in three main groups: constant, time-varying and adaptive inertia weights. A new adaptive inertia weight approach is also proposed which uses the success rate of the swarm as its feedback parameter to ascertain the particles' situation in the search space. The empirical studies on fifteen static test problems, a dynamic function and a real world engineering problem show that the proposed particle swarm optimization model is quite effective in adapting the value of w in the dynamic and static environments.