Abstract: Put succinctly, the bag-of-words based image representation is the most successful approach for object and scene recognition. Within the bag-of-words framework the optimal fusion of multiple cues, such as shape, texture and color, still remains an active research domain. There exist two main approaches to combine color and shape information within the bag-of-words framework. The first approach called, early fusion, fuses color and shape at the feature level as a result of which a joint colorshape vocabulary is produced. The second approach, called late fusion, concatenates histogram representation of both color and shape, obtained independently. In the first part of this thesis, we analyze the theoretical implications of both early and late feature fusion. We demonstrate that both these approaches are suboptimal for a subset of object categories. Consequently, we propose a novel method for recognizing object categories when using multiple cues by separately processing the shape and color cues and combining them by modulating the shape features by category specific color attention. Color is used to compute bottom-up and top-down attention maps. Subsequently, the color attention maps are used to modulate the weights of the shape features. Shape features are given more weight in regions with higher attention and vice versa. The approach is tested on several benchmark object recognition data sets and the results clearly demonstrate the effectiveness of our proposed method. In the second part of the thesis, we investigate the problem of obtaining compact spatial pyramid representations for object and scene recognition. Spatial pyramids have been successfully applied to incorporate spatial information into bag-of-words based image representation. However, a major drawback of spatial pyramids is that it leads to high dimensional image representations. We present a novel framework for obtaining compact pyramid representation. The approach reduces the size of a high dimensional pyramid representation upto an order of magnitude without any significant reduction in accuracy. Moreover, we also investigate the optimal combination of multiple features such as color and shape within the context of our compact pyramid representation. Finally, we describe a novel technique to build discriminative visual words from multiple cues learned independently from training images. To this end, we use an information theoretic vocabulary compression technique to find discriminative combinations of visual cues and the resulting visual vocabulary is compact, has the cue binding property, and supports individual weighting of cues in the final image representation. The approach is tested on standard object recognition data sets. The results obtained clearly demonstrate the effectiveness of our approach.

Abstract: We describe a novel technique for feature combination in the bag-of-words model of image classification. Our approach builds discriminative compound words from primitive cues learned independently from training images. Our main observation is that modeling joint-cue distributions independently is more statistically robust for typical classification problems than attempting to empirically estimate the dependent, joint-cue distribution directly. We use Information theoretic vocabulary compression to find discriminative combinations of cues and the resulting vocabulary of portmanteau words is compact, has the cue binding property, and supports individual weighting of cues in the final image representation. State-of-the-art results on both the Oxford Flower-102 and Caltech-UCSD Bird-200 datasets demonstrate the effectiveness of our technique compared to other, significantly more complex approaches to multi-cue image representation

Abstract: The visual system perceives the colour of surfaces approximately constant under changes of illumination. In this work, we investigate how stable is the perception of categorical \“focal\” colours and their interrelations with varying illuminants and simple chromatic backgrounds. It has been proposed that best examples of colour categories across languages cluster in small regions of the colour space and are restricted to a set of 11 basic terms (Kay and Regier, 2003 Proceedings of the National Academy of Sciences of the USA 100 9085\–9089). Following this, we developed a psychophysical paradigm that exploits the ability of subjects to reliably reproduce the most representative examples of each category, adjusting multiple test patches embedded in a coloured Mondrian. The experiment was run on a CRT monitor (inside a dark room) under various simulated illuminants. We modelled the recorded data for each subject and adapted state as a 3D interconnected structure (graph) in Lab space. The graph nodes were the subject\’s focal colours at each adaptation state. The model allowed us to get a better distance measure between focal structures under different illuminants. We found that perceptual focal structures tend to be preserved better than the structures of the physical \“ideal\” colours under illuminant changes.

Abstract: In the last years, intrinsic image decomposition has gained attention. Most of the state-of-the-art methods are based on the assumption that reflectance changes come along with strong image edges. Recently, user intervention in the recovery problem has proved to be a remarkable source of improvement. In this paper, we propose a novel approach that aims to overcome the shortcomings of pure edge-based methods by introducing strong surface descriptors, such as the color-name descriptor which introduces high-level considerations resembling top-down intervention. We also use a second surface descriptor, termed color-shade, which allows us to include physical considerations derived from the image formation model capturing gradual color surface variations. Both color cues are combined by means of a Markov Random Field. The method is quantitatively tested on the MIT ground truth dataset using different error metrics, achieving state-of-the-art performance.

Abstract: We propose a method to acquire multispectral data with handheld devices with front-mounted RGB cameras. We propose to use the display of the device as an illuminant while the camera captures images illuminated by the red, green and
blue primaries of the display. Three illuminants and three response functions of the camera lead to nine response values which are used for reflectance estimation. Results are promising and show that the accuracy of the spectral reconstruction improves in the range from 30-40% over the spectral
reconstruction based on a single illuminant. Furthermore, we propose to compute sensor-illuminant aware linear basis by discarding the part of the reflectances that falls in the sensorilluminant null-space. We show experimentally that optimizing reflectance estimation on these new basis functions decreases
the RMSE significantly over basis functions that are independent to sensor-illuminant. We conclude that, multispectral data acquisition is potentially possible with consumer hand-held devices such as tablets, mobiles, and laptops, opening up applications which are currently considered to be unrealistic.

Abstract: Color description is a challenging task because of large variations in RGB values which occur due to scene accidental events, such as shadows, shading, specularities, illuminant color changes, and changes in viewing geometry. Traditionally, this challenge has been addressed by capturing the variations in physics-based models, and deriving invariants for the undesired variations. The drawback of this approach is that sets of distinguishable colors in the original color space are mapped to the same value in the photometric invariant space. This results in a drop of discriminative power of the color description. In this paper we take an information theoretic approach to color description. We cluster color values together based on their discriminative power in a classification problem. The clustering has the explicit objective to minimize the drop of mutual information of the final representation. We show that such a color description automatically learns a certain degree of photometric invariance. We also show that a universal color representation, which is based on other data sets than the one at hand, can obtain competing performance. Experiments show that the proposed descriptor outperforms existing photometric invariants. Furthermore, we show that combined with shape description these color descriptors obtain excellent results on four challenging datasets, namely, PASCAL VOC 2007, Flowers-102, Stanford dogs-120 and Birds-200.

Abstract: Comic books constitute an important cultural heritage asset in many countries. Digitization combined with subsequent comic book understanding would enable a variety of new applications, including content-based retrieval and content retargeting. Document understanding in this domain is challenging as comics are semi-structured documents, combining semantically important graphical and textual parts. Few studies have been done in this direction. In this work we detail a novel approach for closed and non-closed speech balloon localization in scanned comic book pages, an essential step towards a fully automatic comic book understanding. The approach is compared with existing methods for closed balloon localization found in the literature and results are presented.

Abstract: Different studies have shown that languages from industrialized cultures
share a set of 11 basic colour terms: red, green, blue, yellow, pink, purple, brown, orange, black, white, and grey (Berlin & Kay, 1969, Basic Color Terms, University of California Press)( Kay & Regier, 2003, PNAS, 100, 9085-9089). Some of these studies have also reported the best representatives or focal values of each colour (Boynton and Olson, 1990, Vision Res. 30,1311–1317), (Sturges and Whitfield, 1995, CRA, 20:6, 364–376). Some further studies have provided us with fuzzy datasets for color naming by asking human observers to rate colours in terms of membership values (Benavente -et al-, 2006, CRA. 31:1, 48–56,). Recently, a computational model based on these human ratings has been developed (Benavente -et al-, 2008, JOSA-A, 25:10, 2582-2593). This computational model follows a fuzzy approach to assign a colour name to a particular RGB value. For example, a pixel with a value (255,0,0) will be named 'red' with membership 1, while a cyan pixel with a RGB value of (0, 200, 200) will be considered to be 0.5 green and 0.5 blue. In this work, we show how this colour naming paradigm can be applied to different computer vision tasks. In particular, we report results in colour constancy (Vazquez-Corral -et al-, 2012, IEEE TIP, in press) showing that the classical constraints on either illumination or surface reflectance can be substituted by
the statistical properties encoded in the colour names. [Supported by projects TIN2010-21771-C02-1, CSD2007-00018].