From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

We study the problem of object classification when training and test classes are disjoint, i.e. no training examples of the target classes are available. This setup has hardly been studied in computer vision research, but it is the rule rather than the exception, because the world contains tens of thousands of different object classes and for only a very few of them image, collections have been formed and annotated with suitable class labels. In this paper, we tackle the problem by introducing attribute-based classification. It performs object detection based on a human-specified high-level description of the target objects instead of training images. The description consists of arbitrary semantic attributes, like shape, color or even geographic information. Because such properties transcend the specific learning task at hand, they can be pre-learned, e.g. from image datasets unrelated to the current task. Afterwards, new classes can be detected based on their attribute representation, without the need for a new training phase. In order to evaluate our method and to facilitate research in this area, we have assembled a new large-scale dataset, “Animals with Attributes”, of over 30,000 animal images that match the 50 classes in Osherson's classic table of how strongly humans associate 85 semantic attributes with animal classes. Our experiments show that by using an attribute layer it is indeed possible to build a learning object detection system that does not require any training images of the target classes.

/pdf/learning-to-detect-unseen-object-classes-by-between-class-nn6vwlm4s2.pdf

Learning to detect unseen object classes by between-class attribute transfer

A classic photographic task is the mapping of the potentially high dynamic range of real world luminances to the low dynamic range of the photographic print. This tone reproduction problem is also faced by computer graphics practitioners who map digital images to a low dynamic range print or screen. The work presented in this paper leverages the time-tested techniques of photographic practice to develop a new tone reproduction operator. In particular, we use and extend the techniques developed by Ansel Adams to deal with digital images. The resulting algorithm is simple and produces good results for a wide variety of images.

/pdf/photographic-tone-reproduction-for-digital-images-k89wmpv97f.pdf

Photographic tone reproduction for digital images

We study the problem of object recognition for categories for which we have no training examples, a task also called zero--data or zero-shot learning. This situation has hardly been studied in computer vision research, even though it occurs frequently; the world contains tens of thousands of different object classes, and image collections have been formed and suitably annotated for only a few of them. To tackle the problem, we introduce attribute-based classification: Objects are identified based on a high-level description that is phrased in terms of semantic attributes, such as the object's color or shape. Because the identification of each such property transcends the specific learning task at hand, the attribute classifiers can be prelearned independently, for example, from existing image data sets unrelated to the current task. Afterward, new classes can be detected based on their attribute representation, without the need for a new training phase. In this paper, we also introduce a new data set, Animals with Attributes, of over 30,000 images of 50 animal classes, annotated with 85 semantic attributes. Extensive experiments on this and two more data sets show that attribute-based classification indeed is able to categorize images without access to any training images of the target classes.

Attribute-Based Classification for Zero-Shot Visual Object Categorization

Distributional semantic models derive computational representations of word meaning from the patterns of co-occurrence of words in text. Such models have been a success story of computational linguistics, being able to provide reliable estimates of semantic relatedness for the many semantic tasks requiring them. However, distributional models extract meaning information exclusively from text, which is an extremely impoverished basis compared to the rich perceptual sources that ground human semantic knowledge. We address the lack of perceptual grounding of distributional models by exploiting computer vision techniques that automatically identify discrete "visual words" in images, so that the distributional representation of a word can be extended to also encompass its co-occurrence with the visual words of images it is associated with. We propose a flexible architecture to integrate text- and image-based distributional information, and we show in a set of empirical tests that our integrated model is superior to the purely text-based approach, and it provides somewhat complementary semantic information with respect to the latter.

/pdf/multimodal-distributional-semantics-3789nky6kq.pdf

Multimodal distributional semantics

We asked human observers to adjust the color of natural fruit objects until they appeared achromatic. The objects were generally perceived to be gray when their color was shifted away from the observers' gray point in a direction opposite to the typical color of the fruit. These results show that color sensations are not determined by the incoming sensory data alone, but are significantly modulated by high-level visual memory.

Memory modulates color appearance.

People perceive roughly constant surface colors despite large changes in illumination. The familiarity of colors of some natural objects might help achieve this feat through direct modulation of the objects' color appearance. Research on memory colors and color appearance has yielded controversial results and due to the employed methods has often confounded perceptual with semantic effects. We studied the effect of memory colors on color appearance by presenting photographs of fruit on a monitor under various simulated illuminations and by asking observers to make either achromatic or typical color settings without placing demands on short-term memory or semantic processing. In a control condition, we presented photographs of 3D fruit shapes without texture and 2D outline shapes. We found that (1) achromatic settings for fruit were systematically biased away from the gray point toward the opposite direction of a fruit's memory color; (2) the strength of the effect depended on the degree of naturalness of the stimuli; and (3) the effect was evident under all tested illuminations, being strongest for illuminations whose chromaticity was closest to the stimulus chromaticity. We conclude that the visual identity of an object has a measurable effect on color perception, and that this effect is robust under illuminant changes, indicating its potential significance as an additional mechanism for color constancy.

/pdf/color-appearance-of-familiar-objects-effects-of-object-shape-2pnuvosvog.pdf

Color appearance of familiar objects: effects of object shape, texture, and illumination changes.

Color perception changes across the visual field. It is best in the fovea and declines in the periphery. Sensitivity to red-green color variations declines more steeply toward the periphery than sensitivity to luminance or blue-yellow colors. It is thought that this decline is due to the increasing size of receptive fields of parvocellular retinal ganglion cells and the unselective or random contribution of L- and M-cones to the receptive field surround. In earlier psychophysical studies it has been found that L - M cone opponency becomes absent above 30 deg. However, physiological experiments in macaque monkeys have shown that midget ganglion cells exist in the intermediate zone of the peripheral retina (20-50 deg) that are strongly cone opponent. Here we explore this contradiction between physiological and psychophysical research, using stimuli of variable size at eccentricities of up to 50 deg. We found that chromatic detection gets worse with increasing eccentricity but is still possible even at large eccentricities. Our results show that chromatic detection at these eccentricities is mediated by cone-opponent mechanisms.

/pdf/color-perception-in-the-intermediate-periphery-of-the-visual-161bg56ta5.pdf

Color perception in the intermediate periphery of the visual field.

We investigated the memory colour effect for colour diagnostic artificial objects. Since knowledge about these objects and their colours has been learned in everyday life, these stimuli allow the investigation of the influence of acquired object knowledge on colour appearance. These investigations are relevant for questions about how object and colour information in high-level vision interact as well as for research about the influence of learning and experience on perception in general. In order to identify suitable artificial objects, we developed a reaction time paradigm that measures (subjective) colour diagnosticity. In the main experiment, participants adjusted sixteen such objects to their typical colour as well as to grey. If the achromatic object appears in its typical colour, then participants should adjust it to the opponent colour in order to subjectively perceive it as grey. We found that knowledge about the typical colour influences the colour appearance of artificial objects. This effect was particularly strong along the daylight axis.

/pdf/object-knowledge-modulates-colour-appearance-45t2nvxhc0.pdf

Object Knowledge Modulates Colour Appearance

Form vision is traditionally regarded as processing primarily achromatic information. Previous investigations into the statistics of color and luminance in natural scenes have claimed that luminance and chromatic edges are not independent of each other and that any chromatic edge most likely occurs together with a luminance edge of similar strength. Here we computed the joint statistics of luminance and chromatic edges in over 700 calibrated color images from natural scenes. We found that isoluminant edges exist in natural scenes and were not rarer than pure luminance edges. Most edges combined luminance and chromatic information but to varying degrees such that luminance and chromatic edges were statistically independent of each other. Independence increased along successive stages of visual processing from cones via postreceptoral color-opponent channels to edges. The results show that chromatic edge contrast is an independent source of information that can be linearly combined with other cues for the proper segmentation of objects in natural and artificial vision systems. Color vision may have evolved in response to the natural scene statistics to gain access to this independent information.

/pdf/independence-of-color-and-luminance-edges-in-natural-scenes-3uycg0xohl.pdf

Thorsten Hansen

Papers

Memory modulates color appearance.

Color appearance of familiar objects: effects of object shape, texture, and illumination changes.

Color perception in the intermediate periphery of the visual field.

Object Knowledge Modulates Colour Appearance

Independence of color and luminance edges in natural scenes