Search

Full-reference image quality measures are a fundamental tool to approximate the human visual system in various applications for digital data management: from retrieval to compression to detection of unauthorized uses. Inspired by both the effectiveness and the simplicity of hand-crafted Structural Similarity Index Measure (SSIM), in this work, we present a framework for the formulation of SSIM-like image quality measures through genetic programming. We explore different terminal sets, defined from the building blocks of structural similarity at different levels of abstraction, and we propose a two-stage genetic optimization that exploits hoist mutation to constrain the complexity of the solutions. Our optimized measures are selected through a cross-dataset validation procedure, which results in superior performance against different versions of structural similarity, measured as correlation with human mean opinion scores. We also demonstrate how, by tuning on specific datasets, it is possible to obtain solutions that are competitive with (or even outperform) more complex image quality measures.

We present a general-purpose framework for the optimization of parametric contrast enhancement algorithms. We first define a regression module for image acceptability, which is based on deep neural features and which is trained on a large dataset of user-expressed preferences. This regression module is then used as the objective function of a Bayesian optimization process, guiding the search for the optimal parameters of a given contrast enhancement algorithm. In our experiments we optimize three different contrast enhancement algorithms of varying levels of complexity. The effectiveness of our optimization framework is experimentally confirmed by evaluating the output of the optimized contrast enhancement algorithms with respect to reference enhanced images.

We propose a network architecture to perform efficient scene understanding. This work presents three main novelties: the first is a module named Improved Guided Upsampling Module that can replace in toto the decoder part in common semantic segmentation networks. Our second contribution is the introduction of a new module based on spatial sampling to perform Instance Segmentation. It provides a very fast instance segmentation needing only a simple post-processing step at inference time. Finally, we propose a novel efficient network design that includes the new modules and test it against different datasets for outdoor scene understanding. To our knowledge, our network is one of the most efficient architectures for scene understanding published to date, furthermore being 8.6% more accurate than the fastest competitor on semantic segmentation and almost five times faster than the most efficient network for instance segmentation.

We present a state of the art method for vegetable and fruit recognition based on convolutional neural networks. We developed our solution around the concept of a smart kitchen/refrigerator equipped with an on-board camera. With this objective in mind, we adopted a dataset that was specifically collected and annotated according to the eating characteristics of the portrayed items. We performed two types of experiment: we first trained and evaluated different state-of-the-art neural architectures on the task of vegetable and fruit recognition. Secondly, we designed and tested a solution that exploits the hierarchical nature of such classes to further improve the final performance of our system. Experimental results demonstrate the quantitative superiority of the proposed solution compared to existing approaches.

In this paper we present a deep learning method to estimate the illuminant of an image. Our model is not trained with illuminant annotations, but with the objective of improving performance on an auxiliary task such as object recognition. To the best of our knowledge, this is the first example of a deep learning architecture for illuminant estimation that is trained without ground truth illuminants. We evaluate our solution on standard datasets for color constancy, and compare it with state of the art methods. Our proposal is shown to outperform most deep learning methods in a cross-dataset evaluation setup, and to present competitive results in a comparison with parametric solutions.

In this paper we present the UMB-DB 3D face database. The database has been built to test algorithms and systems for 3D face analysis in uncontrolled and challenging scenarios, in particular in those cases where faces are occluded. The database is composed of 1473 pairs of depth and color images of 143 subjects. Each subject has been acquired with different facial expressions, and with the face partially occluded by various objects such as eyeglasses, hats, scarves and hands. The total number of occluded acquisitions is 578. The database, that is freely available for research purposes, could be used for various investigations, some of which are suggested in the paper. For the sake of comparison, we report the results of some of the 3D face detection and recognition algorithms in the state of the art. © 2011 IEEE.

In this paper we present a preliminary study towards the design of a low cost colorimetric imaging device. We show that an effective color measurement is possible even using consumer camera sensors. We present here the colorimetric accuracy test results obtained with simulated data using the theoretical model of image formation and a camera simulation system. The final paper will also include the colorimetric accuracy test results obtained with real camera sensors. For all the experiments the multispectral response has been obtained by the overlap of colored filters on the existing color filter array pattern. Many consumer devices can benefit from the colorimetric accuracy of such a system, e.g. printers with automatic color management characterization or other consumer devices where an accurate color response is required. © 2011 IEEE.

One of the most complete techniques that can be used to digitize an artefact is to generate its photo-textured 3D model, which combines high precision metrical information with a faithful color description of the object surfaces. In this work we focus on how to obtain reliable color information by colorimetrically characterizing the color sensors of the imaging device. To this end, a novel target based characterization procedure is proposed that exploits empirical polynomial modeling for colorimetric data estimation. Experimental results are reported and discussed. ©2009 IEEE.

n this paper we consider combining strategies of several white balance algorithms available in the literature to improve the illuminant chromaticity estimation and correction for digital images. We have tested and compared the original algorithms and the combining strategies on three databases of carefully controlled real data, composed respectively of multispectral, CFA and RGB images. The experimental results, evaluated using the Wilcoxon Sign Test, are reported and commented.

This paper presents an innovative restoration strategy which allows for an effective recognition of 3D faces, even when they are partially occluded by unforeseen, extraneous objects such as scarves, hats, glasses, and so on. First, the occluded regions are detected by considering their effects on the projections of the faces in a suitable face space; the non-occluded regions are then used to restore the missing information. Any recognition algorithm can be applied as usual to restored faces. This restoration strategy led to very satisfactory results on a test set of 52 three-dimensional faces presenting various kinds of occlusions

Visual summaries allow users to grasp the overall content of a video, and access specific sequences within the video itself. The post-processing method presented in this paper makes it possible create exhaustive and not redundant visual video summaries. Three steps compose the method: the first removes the meaningless key frames; the second groups the key frames into clusters to allow multi-level summary presentation; and the third identifies the default summary level. © 2006 IEEE.

Three-dimensional data generated from range scanners is usually composed of a huge amount of information. Simplification and compression techniques must be adopted in order to reduce transmission or processing time and to allow real-time visualization. In this paper we propose an approach to semantic simplification of triangular meshes representing faces. The algorithm is aimed to preserve facial features and can be used especially for face recognition systems purposes. In a first phase we detect salient regions using a 3D face detector based on curvature analysis and holistic classification. In a second phase, vertex decimation is applied to the mesh with different decimation parameters for salient and non salient-regions. We have tested our algorithm on a set of 150 acquisitions obtaining good visual quality meshes with approximately 90% or more of decimated vertexes

We introduce a software architecture for embedded, real-time monitoring systems. The architecture is based both upon sound software engineering principles and industrial experience in the application domain.1 © 2005 IEEE.

We propose a new approach to select the representative frames for video summarization. The representative frames are selected based on the results of the analysis of the events depicted in the shot in terms of Regions of Interest (ROIs). These ROIs are obtained from a biologically based computational model of visual attention. To select the video frames, that will be part of the final visual summary, we exploit an adaptive temporal sampling method that analyzes the visual feature distribution of the ROIs. Preliminary results are presented and discussed.