I M Avatar: Implicit Morphable Head Avatars from Videos Zheng, Yufeng; Abrevaya, Victoria Fernandez; Buhler, Marcel C. ...
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR),
2022-June
Conference Proceeding
Traditional 3D morphable face models (3DMMs) provide fine-grained control over expression but cannot easily capture geometric and appearance details. Neural volumetric representations approach ...photorealism but are hard to animate and do not generalize well to unseen expressions. To tackle this problem, we propose IMavatar (Implicit Morphable avatar), a novel method for learning implicit head avatars from monocular videos. Inspired by the fine-grained control mechanisms afforded by conventional 3DMMs, we represent the expression- and pose-related deformations via learned blendshapes and skinning fields. These attributes are pose-independent and can be used to morph the canonical geometry and texture fields given novel expression and pose parameters. We employ ray marching and iterative root-finding to locate the canonical surface intersection for each pixel. A key contribution is our novel analytical gradient formulation that enables end-to-end training of IMavatars from videos. We show quantitatively and qualitatively that our method improves geometry and covers a more complete expression space compared to state-of-the-art methods. Code and data can be found at https://ait.ethz.ch/projects/2022/IMavatar/.
VariTex: Variational Neural Face Textures Buhler, Marcel C.; Meka, Abhimitra; Li, Gengyan ...
2021 IEEE/CVF International Conference on Computer Vision (ICCV),
2021-Oct.
Conference Proceeding
Deep generative models can synthesize photorealistic images of human faces with novel identities. However, a key challenge to the wide applicability of such techniques is to provide independent ...control over semantically meaningful parameters: appearance, head pose, face shape, and facial expressions. In this paper, we propose VariTex - to the best of our knowledge the first method that learns a variational latent feature space of neural face textures, which allows sampling of novel identities. We combine this generative model with a parametric face model and gain explicit control over head pose and facial expressions. To generate complete images of human heads, we propose an additive decoder that adds plausible details such as hair. A novel training scheme enforces a pose-independent latent space and in consequence, allows learning a one-to-many mapping between latent codes and pose-conditioned exterior regions. The resulting method can generate geometrically consistent images of novel identities under fine-grained control over head pose, face shape, and facial expressions. This facilitates a broad range of downstream tasks, like sampling novel identities, changing the head pose, expression transfer, and more.
A unique challenge in creating high-quality animatable and relightable 3D avatars of real people is modeling human eyes, particularly in conjunction with the surrounding periocular face region. The ...challenge of synthesizing eyes is multifold as it requires 1) appropriate representations for the various components of the eye and the periocular region for coherent viewpoint synthesis, capable of representing diffuse, refractive and highly reflective surfaces, 2) disentangling skin and eye appearance from environmental illumination such that it may be rendered under novel lighting conditions, and 3) capturing eyeball motion and the deformation of the surrounding skin to enable re-gazing. These challenges have traditionally necessitated the use of expensive and cumbersome capture setups to obtain high-quality results, and even then, modeling of the full eye region holistically has remained elusive. We present a novel geometry and appearance representation that enables high-fidelity capture and photorealistic animation, view synthesis and relighting of the eye region using only a sparse set of lights and cameras. Our hybrid representation combines an explicit parametric surface model for the eyeball surface with implicit deformable volumetric representations for the periocular region and the interior of the eye. This novel hybrid model has been designed specifically to address the various parts of that exceptionally challenging facial area - the explicit eyeball surface allows modeling refraction and high frequency specular reflection at the cornea, whereas the implicit representation is well suited to model lower frequency skin reflection via spherical harmonics and can represent non-surface structures such as hair (i.e. eyebrows) or highly diffuse volumetric bodies (i.e. sclera), both of which are a challenge for explicit surface models. Tightly integrating the two representations in a joint framework allows controlled photoreal image synthesis and joint optimization of both the geometry parameters of the eyeball and the implicit neural network in continuous 3D space. We show that for high-resolution close-ups of the human eye, our model can synthesize high-fidelity animated gaze from novel views under unseen illumination conditions, allowing to generate visually rich eye imagery.
Expert Estimates for Feature Relevance are Imperfect de Boer, Patrick M.; Buhler, Marcel C.; Bernstein, Abraham
2017 IEEE International Conference on Data Science and Advanced Analytics (DSAA),
2017-Oct.
Conference Proceeding
An early step in the knowledge discovery process is deciding on what data to look at when trying to predict a given target variable. Most of KDD so far is focused on the workflow after data has been ...obtained, or settings where data is readily available and easily integrable for model induction. However, in practice, this is rarely the case, and many times data requires cleaning and transformation before it can be used for feature selection and knowledge discovery. In such environments, it would be costly to obtain and integrate data that is not relevant to the predicted target variable. To reduce the risk of such scenarios in practice, we often rely on experts to estimate the value of potential data based on its meta information (e.g. its description). However, as we will find in this paper, experts perform abysmally at this task. We therefore developed a methodology, KrowDD, to help humans estimate how relevant a dataset might be based on such meta data. We evaluate KrowDD on 3 real-world problems and compare its relevancy estimates with data scientists' and domain experts'. Our findings indicate large possible cost savings when using our tool in bias-free environments, which may pave the way for lowering the cost of classifier design in practice.
Benefits and harms of tenofovir disoproxil fumarate (TDF) in HIV-infected, antiretroviral treatment (ART)-naïve patients of any age have not been systematically reviewed since recent milestone trials ...were published.
We searched MEDLINE, EMBASE, CENTRAL, SCI, LILACS, WHO GHL, and ClinicalTrials.gov for randomized controlled trials (RCTs) comparing TDF-based treatments with any other ART-regimen (last search 01/2015). Trial characteristics and results were extracted, risks of bias systematically assessed, and treatment effects synthesized in meta-analyses using random-effects models.
We included 22 RCTs (8297 patients). We found no differences between groups for mortality, AIDS, fractures, CD4 cell count, and virological failure; and inconclusive information due to inadequate reporting for cardiovascular events, renal failure, proteinuria, rash, and quality of life. Tenofovir disoproxil fumarate-based regimens significantly reduced total cholesterol (mean difference -18.42 mg/dl; 95% confidence interval CI -22.80 to -14.0), LDL-cholesterol (-9.53 mg/dl; -12.16 to -6.89), HDL-cholesterol (-2.97 mg/dl; -4.41 to -1.53), and triglycerides (-29.77 mg/dl; -38.61 to -20.92), bone mineral density (BMD) (hip: -1.41%; -1.87 to -0.94), and glomerular filtration rate (eGFR) (-3.47 ml/minute; -5.89 to -1.06) over 48 weeks of follow-up. Effects were similar in trials comparing fixed-dose TDF/FTC-based regimens with ABC/3TC-based regimens. We found no influence of baseline viral load on virological failure.
Moderate-quality evidence suggests similar effects of TDF-based treatment regimens and other ART on virological failure. Tenofovir disoproxil fumarate-based regimens are associated with a more favorable lipid profile, but with increased risk of reduced BMD and eGFR. Improved reporting quality is vital to allow assessment of clinical outcomes in future trials.
Endothelin-1 is a 21-amino acid endothelial vasoconstrictor peptide that may be the physiological antagonist of endothelium-derived relaxing factor (EDRF). Endothelin-1 (10(-11)-3 x 10(-7) M) evoked ...potent contractions of isolated internal mammary arteries, internal mammary veins, and saphenous veins, which were enhanced in internal mammary veins as compared with internal mammary arteries (concentration shift, 6.3-fold; p less than 0.05) but not in the saphenous veins. Endothelial removal augmented the response to the peptide (at 3 x 10(-7) M) in internal mammary arteries (p less than 0.05) but not in veins. In the artery, EDRF released by acetylcholine or bradykinin reversed endothelin-1-induced contractions; in saphenous veins, both agonists were much less effective compared with the artery and veins contracted with norepinephrine (p less than 0.005-0.01). This inhibition of endothelium-dependent relaxations in veins occurred at half-maximal contractions but was most prominent at maximal contractions to the peptide. Nitric oxide similarly inhibited contractions to endothelin-1 and norepinephrine in internal mammary arteries, whereas in veins that were contracted with endothelin-1 but not with norepinephrine, the relaxations were blunted (p less than 0.005). The nitric oxide donor SIN-1 and sodium nitroprusside induced complete relaxations of internal mammary arteries but were less effective in veins contracted with endothelin-1 (p less than 0.005). Thus, in normal human arteries, EDRF inhibits endothelin-1-induced contractions, whereas the peptide specifically attenuates the effects of EDRF and nitrovasodilators in veins. This may be important in pathological conditions associated with increased levels of endothelin-1 and in veins used as coronary bypass grafts.