Facial Caricature Research Articles

Matching caricatures to photographs

Facial caricatures are informative funny images that allow us to identify a subject even with a few lines and dots. Matching caricatures to photographs is a challenging cross-modal face matching problem. This paper addresses this problem by defining a set of qualitative face features both for caricatures and photographs where features are automatically extracted from photos and manually labeled in caricatures. Additionally we release a publicly available caricature-photograph database with 200 caricatures and corresponding photomates. In our experiments, we use genetic algorithms and logistic regression and achieve over $$\mathrm{30.0}\,\%$$ recognition rate at 0.1 false accept rate.

Neural correlates of facilitations in face learning by selective caricaturing of facial shape or reflectance

Spatially caricatured faces were recently shown to benefit face learning (Schulz et al., 2012a). Moreover, spatial information may be particularly important for encoding unfamiliar faces, but less so for recognizing familiar faces (Kaufmann et al., 2013). To directly test the possibility of a major role of reflectance information for the recognition of familiar faces, we compared effects of selective photorealistic caricaturing in either shape or reflectance on face learning and recognition. Participants learned 3D-photographed faces across different viewpoints, and different images were presented at learning and test. At test, performance benefits for both types of caricatures were modulated by familiarity: Benefits for learned faces were substantially larger for reflectance caricatures, whereas benefits for novel faces were numerically larger for shape caricatures. ERPs confirmed a consistent reduction of the occipitotemporal P200 (200–240ms) by shape caricaturing, whereas the most prominent effect of reflectance caricaturing was seen in an enhanced posterior N250 (240–400ms), a component that has been related to the activation of acquired face representations. Our results suggest that performance benefits for face learning caused by distinctive spatial versus reflectance information are mediated by different neural processes with different timing and support a prominent role of reflectance for the recognition of learned faces.

A new theoretical approach to improving face recognition in disorders of central vision: Face caricaturing

Damage to central vision, of which age-related macular degeneration (AMD) is the most common cause, leaves patients with only blurred peripheral vision. Previous approaches to improving face recognition in AMD have employed image manipulations designed to enhance early-stage visual processing (e.g., magnification, increased HSF contrast). Here, we argue that further improvement may be possible by targeting known properties of mid- and/or high-level face processing. We enhance identity-related shape information in the face by caricaturing each individual away from an average face. We simulate early- through late-stage AMD-blur by filtering spatial frequencies to mimic the amount of blurring perceived at approximately 10° through 30° into the periphery (assuming a face seen premagnified on a tablet computer). We report caricature advantages for all blur levels, for face viewpoints from front view to semiprofile, and in tasks involving perceiving differences in facial identity between pairs of people, remembering previously learned faces, and rejecting new faces as unknown. Results provide a proof of concept that caricaturing may assist in improving face recognition in AMD and other disorders of central vision.

Image Based Hair Segmentation Algorithm for the Application of Automatic Facial Caricature Synthesis

Hair is a salient feature in human face region and are one of the important cues for face analysis. Accurate detection and presentation of hair region is one of the key components for automatic synthesis of human facial caricature. In this paper, an automatic hair detection algorithm for the application of automatic synthesis of facial caricature based on a single image is proposed. Firstly, hair regions in training images are labeled manually and then the hair position prior distributions and hair color likelihood distribution function are estimated from these labels efficiently. Secondly, the energy function of the test image is constructed according to the estimated prior distributions of hair location and hair color likelihood. This energy function is further optimized according to graph cuts technique and initial hair region is obtained. Finally, K-means algorithm and image postprocessing techniques are applied to the initial hair region so that the final hair region can be segmented precisely. Experimental results show that the average processing time for each image is about 280 ms and the average hair region detection accuracy is above 90%. The proposed algorithm is applied to a facial caricature synthesis system. Experiments proved that with our proposed hair segmentation algorithm the facial caricatures are vivid and satisfying.

Interactive Evolutionary System for Synthesizing Facial Caricature with Non-planar Expression

Interactive Evolutionary System for Synthesizing Facial Caricature with Non-planar Expression

Simpson Style Caricature based on MLS

We present a novel approach to producing facial caricature with Simpson cartoon style based on Moving Least Squares (MLS). We take advantage of employing the caricature stylization rule of caricature artist, Justin. Our method allows Simpson-style cartoon character similar to user’s features by using Justin’s technique, which is a set of caricature stylization rules. Our method transforms input photo image into Simpson style caricature by using MLS approximation. The unique characteristics of user in the photo can be detected by comparing to the mean face feature and the input face feature extracted by AAM(Active Appearance Model). To exaggerate the detected unique characteristics, we set up the exaggeration rules using Justin’s technique. In addition, during the cartooning process, users hairs and accessories are used to the deformed image to make a close resemblance. Our method preserves the reliable and stylized caricature through the exaggeration rules of the actual caricature artists techniques. From this study, we can easily create a Simpson-style cartoon caricature to resemble users features by combining a caricature with existing cartoon researches.

Understanding the multiframe caricature advantage for recognizing facial composites

Eyewitnesses often construct a “composite” face of a person they saw commit a crime, a picture that police use to identify suspects. We described a technique (Frowd, Bruce, Ross, McIntyre, & Hancock, 2007) based on facial caricature to facilitate recognition of these images: Correct naming substantially improves when composites are seen with progressive positive caricature, where distinctive information is enhanced, and then with progressive negative caricature, the opposite. Over the course of four experiments, the underpinnings of this mechanism were explored. Positive-caricature levels were found to be largely responsible for improving naming of composites, with some benefit from negative-caricature levels. Also, different frame-presentation orders (forward, reverse, random, repeated) facilitated equivalent naming benefit relative to static composites. Overall, the data indicate that composites are usually constructed as negative caricatures.

Faces forming traces: Neurophysiological correlates of learning naturally distinctive and caricatured faces

Distinctive faces are easier to learn and recognise than typical faces. We investigated effects of natural vs. artificial distinctiveness on performance and neural correlates of face learning. Spatial caricatures of initially non-distinctive faces were created such that their rated distinctiveness matched a set of naturally distinctive faces. During learning, we presented naturally distinctive, caricatured, and non-distinctive faces for later recognition among novel faces, using different images of the same identities at learning and test. For learned faces, an advantage in performance was observed for naturally distinctive and caricatured over non-distinctive faces, with larger benefits for naturally distinctive faces. Distinctive and caricatured faces elicited more negative occipitotemporal ERPs (P200, N250) and larger centroparietal positivity (LPC) during learning. At test, earliest distinctiveness effects were again seen in the P200. In line with recent research, N250 and LPC were larger for learned than for novel faces overall. Importantly, whereas left hemispheric N250 was increased for learned naturally distinctive faces, right hemispheric N250 responded particularly to caricatured novel faces. We conclude that natural distinctiveness induces benefits to face recognition beyond those induced by exaggeration of a face's idiosyncratic shape, and that the left hemisphere in particular may mediate recognition across different images.

Cherished chansonniers

For generations, students of early music have been fascinated and enthralled by 15th-century chansonniers, not least among them the five related manuscripts originating in the Loire Valley (the Copenhagen, Dijon, Laborde, Nivelle de La Chaussée and Wolfenbüttel chansonniers). With their filigree borders, their whimsical initials of birds, fish and caricatured faces, the enticing delicacy of their elegant musical notation and fine script—all accomplished within a seemingly impossibly tiny space—it is small wonder that these hand-held gems have been the objects of sustained attention at least since the 19th century, and probably to their many owners throughout their histories. Jane Alden’s infectious enthusiasm for the chansonniers as material objects shines through the pages of Songs, scribes and society, and will strike a chord with anyone who has experienced the thrill of seeing the books in situ, in exhibitions, or in the printed or digital facsimiles now available of all five manuscripts. (Full details of these facsimiles, including links to the locations of online images, are listed on p.60 of the book, and in the summary information on each manuscript contained within Appendix 1.) Despite their comparatively high profile, however, Alden argues that the materiality of these books, or the ‘dynamic potential’ of the ‘musical, poetic, and artistic traditions contained within these small volumes’, has been unjustly neglected, and sets out on a thorough investigation of the books as objects which aims to provide ‘important insight into the place of song in later 15th-century society’ (p.167).

Mean value coordinates–based caricature and expression synthesis

We present a novel method for caricature synthesis based on mean value coordinates (MVC). Our method can be applied to any single frontal face image to learn a specified caricature face pair for frontal and 3D caricature synthesis. This technique only requires one or a small number of exemplar pairs and a natural frontal face image training set, while the system can transfer the style of the exemplar pair across individuals. Further exaggeration can be fulfilled in a controllable way. Our method is further applied to facial expression transfer, interpolation, and exaggeration, which are applications of expression editing. Additionally, we have extended our approach to 3D caricature synthesis based on the 3D version of MVC. With experiments we demonstrate that the transferred expressions are credible and the resulting caricatures can be characterized and recognized.

Colored exaggerative caricature creation using inter- and intra-correlations of feature shapes and positions

This paper develops a system comprising a statistics-based exaggerative (SBE) module and a non-photorealistic rendering (NPR) module for the automatic creation of colored facial caricatures with exaggerated facial features and individual facial details such as beards and moles. Unlike previous research that focused on the inter-correlation (the difference between the facial features of input image and those of the mean face in the training database), the SBE module exaggerates the input image utilizing an iterative approach based on both inter- and intra-correlations of the facial features. The intra-correlation considered in this study makes the comparison with other features within the same input image, and has the effect of exaggerating the major facial features while simultaneously subduing the visual impact of non-major facial features. The NPR module consists of a black-and-white sketch creation process and a colored facial cartoon creation process. The results of the two processes are combined to generate a colored cartoon-like sketch, which is then warped into a colored exaggerative facial caricature based on the corresponding exaggerative shape and position created by the SBE module. The experimental results demonstrate that the proposed method can emphasize the major characteristics of a face better than previous methods that only considered feature inter-correlation.

Adaptation May Cause Some of the Face Caricature Effect

One of the ways to demonstrate a caricature preference is to ask participants to adjust a face image over a range from anti-caricature to caricature until it shows the best likeness to a specific individual. Since facial adaptation, whereby exposure to a face influences subsequent perception of faces, is rapid, it is possible that adaptation promotes the selection of a caricatured image. We tested whether giving participants a reference average face image, to counteract any adaptation, would reduce the degree of caricature selected for famous faces. Results confirmed a significant decrease but, even without an average, participants chose an anti-caricatured image. These data suggest a role for adaptation in generating caricature preferences while also suggesting such preferences are not inevitable.

Frontal and Semi-Frontal Facial Caricature Synthesis Using Non-Negative Matrix Factorization

In this paper, we present a novel approach to synthesizing frontal and semi-frontal cartoon-like facial caricatures from an image. The caricature is generated by warping the input face from the original feature points to the corresponding exaggerated feature points. A 3D mean face model is incorporated to facilitate face to caricatures by inferring the depth of 3D feature points and the spatial transformation. Then the 3D face is deformed by using non-negative matrix factorization and projected back to image plane for future warping. To efficiently solve the nonlinear spatial transformation, we propose a novel initialization scheme to set up Levenberg-Marquardt optimization. According to the spatial transformation, exaggeration is applied to the most salient features by exaggerating their normalized difference from the mean. Non-photorealistic rendering (NPR) based stylization completes the cartoon caricature. Experiments demonstrate that our method outperforms existing methods in terms of view angles and aesthetic visual quality.

Manifold‐Based 3D Face Caricature Generation with Individualized Facial Feature Extraction

AbstractCaricature is an interesting art to express exaggerated views of different persons and things through drawing. The face caricature is popular and widely used for different applications. To do this, we have to properly extract unique/specialized features of a person's face. A person's facial feature not only depends on his/her natural appearance, but also the associated expression style. Therefore, we would like to extract the neutural facial features and personal expression style for different applicaions. In this paper, we represent the 3D neutral face models in BU–3DFE database by sparse signal decomposition in the training phase. With this decomposition, the sparse training data can be used for robust linear subspace modeling of public faces. For an input 3D face model, we fit the model and decompose the 3D model geometry into a neutral face and the expression deformation separately. The neutral geomertry can be further decomposed into public face and individualized facial feature. We exaggerate the facial features and the expressions by estimating the probability on the corresponding manifold. The public face, the exaggerated facial features and the exaggerated expression are combined to synthesize a 3D caricature for a 3D face model. The proposed algorithm is automatic and can effectively extract the individualized facial features from an input 3D face model to create 3D face caricature.

Cross-modal face identity aftereffects and their relation to priming.

We tested the magnitude of the face identity aftereffect following adaptation to different modes of adaptors in four experiments. The perceptual midpoint between two morphed famous faces was measured pre- and post-adaptation. Significant aftereffects were observed for visual (faces) and nonvisual adaptors (voices and names) but not nonspecific semantic information (e.g., occupations). Aftereffects were also observed following imagination and adaptation to an associated person. The strongest aftereffects were found adapting to facial caricatures. These results are discussed in terms of cross-modal adaptation occurring at various loci within the face-recognition system analogous to priming.

The reverse‐caricature effect revisited: Familiarization with frontal facial caricatures improves veridical face recognition

Prior research suggests that recognition of a person's face can be facilitated by exaggerating the distinctive features of the face during training. We tested if this 'reverse-caricature effect' would be robust to procedural variations that created more difficult learning environments. Specifically, we examined whether the effect would emerge with frontal rather than three-quarter views, after very brief exposure to caricatures during the learning phase and after modest rotations of faces during the recognition phase. Results indicate that, even under these difficult training conditions, people are more accurate at recognizing unaltered faces if they are first familiarized with caricatures of the faces, rather than with the unaltered faces. These findings support the development of new training methods to improve face recognition.

A NEW METHOD FOR FACIAL FEATURES QUANTIFICATION OF CARICATURE BASED ON SELF-REFERENCE MODEL

Some facial features that differ from an ordinary face should be identified by a computer when generating a facial caricature. These distinctive facial features are called self-features. Compared with traditional Mean Face Model (MFM) that is unable to quantify these self-features well, a Self-Reference Model (SRM) is presented in this paper. Firstly, based on the physiology structure of a front face, a self-reference is found, and this reference is used to measure the self-features. According to the self-reference, some standard facial parameters are worked out by collecting statistic data of many facial images. Then, in an input face image, by evaluating some differences between the input face and the standard facial parameters, the self-features are properly estimated and quantified. Finally, by analyzing some caricatures produced by caricaturists, the SRM can prove the validity of the proposed Algorithm.

A bilateral occipitotemporal network mediates face perception

The aim of the present study was to further explore the neuronal mechanisms of face processing in healthy subjects which may help to understand the difficulties experienced by prosopagnosia subjects. A further goal was to compare face specific activation patterns in the right and left occipital face area (OFA) and fusiform face area (FFA) for famous faces, non-famous faces and caricatures of famous faces in four individuals suffering from developmental prosopagnosia (DP) and seven healthy controls, using functional magnetic resonance imaging and psychophysiological interaction analysis (PPI). Control subjects showed higher face related activations in the right compared to the left FFA. Caricatures of faces of famous people and photographs of non-famous faces yielded higher percent signal changes in the OFA and FFA compared to photographs of famous faces. These data support the idea that the OFA and FFA discriminate between familiar and new face representations. The activation patterns of DP subjects were heterogeneous, with none of the patients showing bilateral face related activations in both OFA and FFA. There was no evidence of a left hemispheric activation when the right homologue failed to be activated, supporting the view of a right hemispheric dominance in face perception. PPI analysis indicated a link between activation of the right FFA and the other three tested regions, the left FFA and the right and left OFA. In summary, all four face related brain regions appear to be necessary for successful face processing, and disruption of one component may lead to face recognition deficits.

Amygdala and fusiform gyrus temporal dynamics: Responses to negative facial expressions

BackgroundThe amygdala habituates in response to repeated human facial expressions; however, it is unclear whether this brain region habituates to schematic faces (i.e., simple line drawings or caricatures of faces). Using an fMRI block design, 16 healthy participants passively viewed repeated presentations of schematic and human neutral and negative facial expressions. Percent signal changes within anatomic regions-of-interest (amygdala and fusiform gyrus) were calculated to examine the temporal dynamics of neural response and any response differences based on face type.ResultsThe amygdala and fusiform gyrus had a within-run "U" response pattern of activity to facial expression blocks. The initial block within each run elicited the greatest activation (relative to baseline) and the final block elicited greater activation than the preceding block. No significant differences between schematic and human faces were detected in the amygdala or fusiform gyrus.ConclusionThe "U" pattern of response in the amygdala and fusiform gyrus to facial expressions suggests an initial orienting, habituation, and activation recovery in these regions. Furthermore, this study is the first to directly compare brain responses to schematic and human facial expressions, and the similarity in brain responses suggest that schematic faces may be useful in studying amygdala activation.

An application of caricature: How to improve the recognition of facial composites

Facial caricatures exaggerate the distinctive features of a face and may elevate the recognition of a familiar face. We investigate whether the recognition of facial composites, or pictures of criminal faces, could be similarly enhanced. In this study, participants first estimated the degree of caricature necessary to make composites most identifiable. Contrary to expectation, an anticaricature was found to be best, presumably as this tended to reduce the appearance of errors. In support of this explanation, more positive caricature estimates were assigned to morphed composites: representations that tend to contain less overall error. In addition, anticaricaturing reduced identification for morphed composites but enhanced identification for individual composites. Although such improvements were too small to be of value to law enforcement, a sizeable naming benefit was observed when presenting a range of caricature states, which appeared to capitalize on individual differences in the internal representation of familiar faces.