Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
30,258
result(s) for
"realistic"
Sort by:
Racing Manhattan
Alone in the world, Jay Barton is a teenage misfit with nothing much going for her besides an extraordinary talent for understanding racehorses and riding them like a pro. When, in a desperate attempt to escape her shifty, opportunistic uncle, she leaves home to work in a racing stable, Jay forms a bond with a beautiful gray mare named Manhattan--brilliant, misunderstood, dangerous, and heading for racing's scrap heap. Recognizing a fellow misfit, Jay fights to give Manhattan one last opportunity to show that she's the champion she was born to be. Together they face a world of prejudice and cruelty, fighting back the only way they know how--by becoming the best.
Book
Realistic Speech-Driven Facial Animation with GANs
Speech-driven facial animation is the process that automatically synthesizes talking characters based on speech signals. The majority of work in this domain creates a mapping from audio features to visual features. This approach often requires post-processing using computer graphics techniques to produce realistic albeit subject dependent results. We present an end-to-end system that generates videos of a talking head, using only a still image of a person and an audio clip containing speech, without relying on handcrafted intermediate features. Our method generates videos which have (a) lip movements that are in sync with the audio and (b) natural facial expressions such as blinks and eyebrow movements. Our temporal GAN uses 3 discriminators focused on achieving detailed frames, audio-visual synchronization, and realistic expressions. We quantify the contribution of each component in our model using an ablation study and we provide insights into the latent representation of the model. The generated videos are evaluated based on sharpness, reconstruction quality, lip-reading accuracy, synchronization as well as their ability to generate natural blinks.
Journal Article
The Effects of Perceived Identity Threat and Realistic Threat on the Negative Attitudes and Usage Intentions Toward Hotel Service Robots: The Moderating Effect of the Robot’s Anthropomorphism
The use of social robots in service scenarios (e.g. in hotels) is expected to increase. Research has indicated that realistic threat and identity threat contribute to prejudice, discrimination, and conflict. Therefore, designing smart robots that can understand user needs and provide prompt service is critical. This study aimed to explore the relationships among perceived threat, negative attitudes toward robots, and usage intention, and to discuss how anthropomorphism moderates the relationship between negative attitudes and usage intention toward robots. We hypothesized that realistic threat and identity threat positively influence negative attitudes toward social robots; such negative attitudes have negative effects on usage intention; and anthropomorphism positively moderates said effect. A between-subjects factorial research design was employed; participants were randomly assigned to one of two conditions of robot anthropomorphic appearance. The stimuli were two pictures of service robots—one with a humanoid appearance and one without. After viewing one of the stimuli, participants filled in a questionnaire that assessed the realistic threat, identity threat, negative attitudes, and usage intention. When participants perceived the robot’s appearance to be highly anthropomorphic, their negative attitudes toward it had a stronger negative effect on their usage intention than when they considered the appearance less anthropomorphic. Both identity threat and realistic threat significantly increased negative attitudes toward the robots; thus, the public’s willingness to use these robots would be negatively affected. This study recommends that robotics companies consider how to decrease people’s perceived realistic and identity threats as well as adjust robots’ anthropomorphic appearance to people’s tastes.
Journal Article
Towards High Fidelity Face Frontalization in the Wild
Face frontalization refers to the process of synthesizing the frontal view of a face from a given profile. Due to self-occlusion and appearance distortion in the wild, it is extremely challenging to recover faithful high-resolution results meanwhile preserve texture details. This paper proposes a high fidelity pose in-variant model (HF-PIM) to produce photographic and identity-preserving results. HF-PIM frontalizes the profiles through a novel texture fusion warping procedure and leverages a dense correspondence field to bind the 2D and 3D surface spaces. We decompose the prerequisite of warping into dense correspondence field estimation and facial texture map recovering, which are both well addressed by deep networks. Different from those reconstruction methods relying on 3D data, we also propose adversarial residual dictionary learning to supervise facial texture map recovering with only monocular images. Furthermore, a multi-perception guided loss is proposed to address the practical misalignment between the ground truth frontal and profile faces, allowing HF-PIM to effectively utilize multiple images during training. Quantitative and qualitative evaluations on five controlled and uncontrolled databases show that the proposed method not only boosts the performance of pose-invariant face recognition but also improves the visual quality of high-resolution frontalization appearances.
Journal Article
LINEARLY DECOUPLED ENERGY-STABLE NUMERICAL METHODS FOR MULTICOMPONENT TWO-PHASE COMPRESSIBLE FLOW
In this paper, for the first time we propose two linear, decoupled, energy-stable numerical schemes for multicomponent two-phase compressible flow with a realistic equation of state (e.g., Peng–Robinson equation of state). The methods are constructed based on the scalar auxiliary variable (SAV) approaches for Helmholtz free energy and the intermediate velocities that are designed to decouple the tight relationship between velocity and molar densities. The intermediate velocities are also involved in the discrete momentum equation to ensure consistency with the mass balance equations. Moreover, we propose a componentwise SAV approach for a multicomponent fluid, which requires solving a sequence of linear, separate mass balance equations. The fully discrete schemes are also constructed based on the finite difference/volume methods with the upwind scheme on staggered grids. We prove that the semidiscrete and fully discrete schemes preserve the unconditional energy-dissipation feature. Numerical results are presented to verify the effectiveness of the proposed methods.
Journal Article
Inter-Subject Variability of Skull Conductivity and Thickness in Calibrated Realistic Head Models
•We propose a non-invasive procedure for estimating skull conductivity and thickness.•We evaluate this procedure with EEG, MEG and MRI data of twenty participants.•A high inter-subject variability is found for skull conductivity and thickness.•Skull conductivity is significantly correlated with age and skull thickness.•This procedure improves head modeling for EEG/MEG source analysis and optimized TES.
Skull conductivity has a substantial influence on EEG and combined EEG and MEG source analysis as well as on optimized transcranial electric stimulation. To overcome the use of standard literature values, we propose a non-invasive two-level calibration procedure to estimate skull conductivity individually in a group study with twenty healthy adults. Our procedure requires only an additional run of combined somatosensory evoked potential and field data, which can be easily integrated in EEG/MEG experiments. The calibration procedure uses the P20/N20 topographies and subject-specific realistic head models from MRI. We investigate the inter-subject variability of skull conductivity and relate it to skull thickness, age and gender of the subjects, to the individual scalp P20/N20 surface distance between the P20 potential peak and the N20 potential trough as well as to the individual source depth of the P20/N20 source. We found a considerable inter-subject variability for (calibrated) skull conductivity (8.44 ± 4.84 mS/m) and skull thickness (5.97 ± 1.19 mm) with a statistically significant correlation between them (rho = 0.52). Age showed a statistically significant negative correlation with skull conductivity (rho = -0.5). Furthermore, P20/N20 surface distance and source depth showed large inter-subject variability of 12.08 ± 3.21 cm and 15.45 ± 4.54 mm, respectively, but there was no significant correlation between them. We also found no significant differences among gender subgroups for the investigated measures. It is thus important to take the inter-subject variability of skull conductivity and thickness into account by means of using subject-specific calibrated realistic head modeling.
Journal Article
Guided Hyperspectral Image Denoising with Realistic Data
The hyperspectral image (HSI) denoising has been widely utilized to improve HSI qualities. Recently, learning-based HSI denoising methods have shown their effectiveness, but most of them are based on synthetic dataset and lack the generalization capability on real testing HSI. Moreover, there is still no public paired real HSI denoising dataset to learn HSI denoising network and quantitatively evaluate HSI methods. In this paper, we mainly focus on how to produce realistic dataset for learning and evaluating HSI denoising network. On the one hand, we collect a paired real HSI denoising dataset, which consists of short-exposure noisy HSIs and the corresponding long-exposure clean HSIs. On the other hand, we propose an accurate HSI noise model which matches the distribution of real data well and can be employed to synthesize realistic dataset. On the basis of the noise model, we present an approach to calibrate the noise parameters of the given hyperspectral camera. Besides, on the basis of observation of high signal-to-noise ratio of mean image of all spectral bands, we propose a guided HSI denoising network with guided dynamic nonlocal attention, which calculates dynamic nonlocal correlation on the guidance information, i.e., mean image of spectral bands, and adaptively aggregates spatial nonlocal features for all spectral bands. The extensive experimental results show that a network learned with only synthetic data generated by our noise model performs as well as it is learned with paired real data, and our guided HSI denoising network outperforms state-of-the-art methods under both quantitative metrics and visual quality.
Journal Article
Benchmarking the Robustness of Semantic Segmentation Models with Respect to Common Corruptions
When designing a semantic segmentation model for a real-world application, such as autonomous driving, it is crucial to understand the robustness of the network with respect to a wide range of image corruptions. While there are recent robustness studies for full-image classification, we are the first to present an exhaustive study for semantic segmentation, based on many established neural network architectures. We utilize almost 400,000 images generated from the Cityscapes dataset, PASCAL VOC 2012, and ADE20K. Based on the benchmark study, we gain several new insights. Firstly, many networks perform well with respect to real-world image corruptions, such as a realistic PSF blur. Secondly, some architecture properties significantly affect robustness, such as a Dense Prediction Cell, designed to maximize performance on clean data only. Thirdly, the generalization capability of semantic segmentation models depends strongly on the type of image corruption. Models generalize well for image noise and image blur, however, not with respect to digitally corrupted data or weather corruptions.
Journal Article
Deep Neural Network Augmentation: Generating Faces for Affect Analysis
This paper presents a novel approach for synthesizing facial affect; either in terms of the six basic expressions (i.e., anger, disgust, fear, joy, sadness and surprise), or in terms of valence (i.e., how positive or negative is an emotion) and arousal (i.e., power of the emotion activation). The proposed approach accepts the following inputs:(i) a neutral 2D image of a person; (ii) a basic facial expression or a pair of valence-arousal (VA) emotional state descriptors to be generated, or a path of affect in the 2D VA space to be generated as an image sequence. In order to synthesize affect in terms of VA, for this person, 600,000 frames from the 4DFAB database were annotated. The affect synthesis is implemented by fitting a 3D Morphable Model on the neutral image, then deforming the reconstructed face and adding the inputted affect, and blending the new face with the given affect into the original image. Qualitative experiments illustrate the generation of realistic images, when the neutral image is sampled from fifteen well known lab-controlled or in-the-wild databases, including Aff-Wild, AffectNet, RAF-DB; comparisons with generative adversarial networks (GANs) show the higher quality achieved by the proposed approach. Then, quantitative experiments are conducted, in which the synthesized images are used for data augmentation in training deep neural networks to perform affect recognition over all databases; greatly improved performances are achieved when compared with state-of-the-art methods, as well as with GAN-based data augmentation, in all cases.
Journal Article
Honey Bee Colony Health in Thiamethoxam‐Treated Sugar Beet Fields: A Field‐Based Case Study
Emergency use of thiamethoxam seed treatments in sugar beet was approved in Germany in 2021, despite EU restrictions on neonicotinoids because of pollinator risks. During the field experiment underlying this case study, residues in bee‐relevant matrices were detected only at very low levels, and conservative exposure modeling indicated no acute or chronic concern for honey bees. Building on these exposure findings, the present analysis examined whether such exposures translated into measurable effects on honey bee colony performance. In Experiment 1, colony development was monitored at two geographically distinct sites across the 2021 sugar beet season. Colonies at both sites exhibited strong seasonal growth. Generalized linear mixed models detected no consistent adverse effects of thiamethoxam treatment on either adult bee populations or brood cell numbers. Although temporal fluctuations and site‐specific variability were evident, treatment effects were not statistically supported, highlighting the importance of multi‐site approaches when assessing pesticide impacts and the need for continued multi‐year evidence under diverse environmental conditions. In Experiment 2, survival of individual workers was evaluated using free‐flying mini‐hives. Mixed‐effects Cox modeling, which accounted for colony variance, found no significant differences in worker longevity between treated and control groups. This indicates no evidence for reduced worker survival under a field‐relevant neonicotinoid exposure scenario. Together, these two complementary experimental approaches show that thiamethoxam seed treatments in sugar beet did not cause consistent adverse effects on honey bee colonies under the tested agricultural conditions. By integrating residue analyses, statistical modeling, and colony‐level monitoring, the study provides ecologically relevant evidence that current agricultural practices with thiamethoxam in sugar beet pose a low apparent risk to honey bee colony health, while underscoring the value of longer‐term and broader‐scale field evaluations. Using multi‐site field trials during Germany's 2021 emergency authorization of thiamethoxam‐treated sugar beet, we combined colony demography and residue analyses with complementary worker survival assays to characterize honey bee exposure under realistic conditions. Across sites, no consistent adverse effects were observed; a late‐season reduction in adult bee numbers at the JKI site numerically exceeded EFSA's 2023 SPG threshold but was not replicated. Residue data indicated minimal bee exposure within sugar beet production systems.
Journal Article