Victoria Fernández Abrevaya

Haiwen Feng, Timo Bolkart, Joachim Tesch et al. · amazon-science

Virtual facial avatars will play an increasingly important role in immersive communication, games and the metaverse, and it is therefore critical that they be inclusive. This requires accurate recovery of the appearance, represented by albedo, regardless of age, sex, or ethnicity. While significant progress has been made on estimating 3D facial geometry, albedo estimation has received less attention. The task is fundamentally ambiguous because the observed color is a function of albedo and lighting, both of which are unknown. We find that current methods are biased towards light skin tones due to (1) strongly biased priors that prefer lighter pigmentation and (2) algorithmic solutions that disregard the light/albedo ambiguity. To address this, we propose a new evaluation dataset (FAIR) and an algorithm (TRUST) to improve albedo estimation and, hence, fairness. Specifically, we create the first facial albedo evaluation benchmark where subjects are balanced in terms of skin color, and measure accuracy using the Individual Typology Angle (ITA) metric. We then address the light/albedo ambiguity by building on a key observation: the image of the full scene -- as opposed to a cropped image of the face -- contains important information about lighting that can be used for disambiguation. TRUST regresses facial albedo by conditioning both on the face region and a global illumination signal obtained from the scene image. Our experimental results show significant improvement compared to state-of-the-art methods on albedo estimation, both in terms of accuracy and fairness. The evaluation benchmark and code will be made available for research purposes at https://trust.is.tue.mpg.de.

Haiwen Feng, Peter Kulits, Shichen Liu et al.

We address the problem of fitting a parametric human body model (SMPL) to point cloud data. Optimization-based methods require careful initialization and are prone to becoming trapped in local optima. Learning-based methods address this but do not generalize well when the input pose is far from those seen during training. For rigid point clouds, remarkable generalization has been achieved by leveraging SE(3)-equivariant networks, but these methods do not work on articulated objects. In this work we extend this idea to human bodies and propose ArtEq, a novel part-based SE(3)-equivariant neural architecture for SMPL model estimation from point clouds. Specifically, we learn a part detection network by leveraging local SO(3) invariance, and regress shape and pose using articulated SE(3) shape-invariant and pose-equivariant networks, all trained end-to-end. Our novel pose regression module leverages the permutation-equivariant property of self-attention layers to preserve rotational equivariance. Experimental results show that ArtEq generalizes to poses not seen during training, outperforming state-of-the-art methods by ~44% in terms of body reconstruction accuracy, without requiring an optimization refinement step. Furthermore, ArtEq is three orders of magnitude faster during inference than prior work and has 97.3% fewer parameters. The code and model are available for research purposes at https://arteq.is.tue.mpg.de.

Shrisha Bharadwaj, Yufeng Zheng, Otmar Hilliges et al.

Our goal is to efficiently learn personalized animatable 3D head avatars from videos that are geometrically accurate, realistic, relightable, and compatible with current rendering systems. While 3D meshes enable efficient processing and are highly portable, they lack realism in terms of shape and appearance. Neural representations, on the other hand, are realistic but lack compatibility and are slow to train and render. Our key insight is that it is possible to efficiently learn high-fidelity 3D mesh representations via differentiable rendering by exploiting highly-optimized methods from traditional computer graphics and approximating some of the components with neural networks. To that end, we introduce FLARE, a technique that enables the creation of animatable and relightable mesh avatars from a single monocular video. First, we learn a canonical geometry using a mesh representation, enabling efficient differentiable rasterization and straightforward animation via learned blendshapes and linear blend skinning weights. Second, we follow physically-based rendering and factor observed colors into intrinsic albedo, roughness, and a neural representation of the illumination, allowing the learned avatars to be relit in novel scenes. Since our input videos are captured on a single device with a narrow field of view, modeling the surrounding environment light is non-trivial. Based on the split-sum approximation for modeling specular reflections, we address this by approximating the pre-filtered environment map with a multi-layer perceptron (MLP) modulated by the surface roughness, eliminating the need to explicitly model the light. We demonstrate that our mesh-based avatar formulation, combined with learned deformation, material, and lighting MLPs, produces avatars with high-quality geometry and appearance, while also being efficient to train and render compared to existing approaches.

Omri Ben-Dov, Pravir Singh Gupta, Victoria Abrevaya et al.

Generative Adversarial Networks (GANs) can produce high-quality samples, but do not provide an estimate of the probability density around the samples. However, it has been noted that maximizing the log-likelihood within an energy-based setting can lead to an adversarial framework where the discriminator provides unnormalized density (often called energy). We further develop this perspective, incorporate importance sampling, and show that 1) Wasserstein GAN performs a biased estimate of the partition function, and we propose instead to use an unbiased estimator; and 2) when optimizing for likelihood, one must maximize generator entropy. This is hypothesized to provide a better mode coverage. Different from previous works, we explicitly compute the density of the generated samples. This is the key enabler to designing an unbiased estimator of the partition function and computation of the generator entropy term. The generator density is obtained via a new type of flow network, called one-way flow network, that is less constrained in terms of architecture, as it does not require a tractable inverse function. Our experimental results show that our method converges faster, produces comparable sample quality to GANs with similar architecture, successfully avoids over-fitting to commonly used datasets and produces smooth low-dimensional latent representations of the training data.

Omri Ben-Dov, Pravir Singh Gupta, Victoria Fernandez Abrevaya et al.

Modern generative models are roughly divided into two main categories: (1) models that can produce high-quality random samples, but cannot estimate the exact density of new data points and (2) those that provide exact density estimation, at the expense of sample quality and compactness of the latent space. In this work we propose LED, a new generative model closely related to GANs, that allows not only efficient sampling but also efficient density estimation. By maximizing log-likelihood on the output of the discriminator, we arrive at an alternative adversarial optimization objective that encourages generated data diversity. This formulation provides insights into the relationships between several popular generative models. Additionally, we construct a flow-based generator that can compute exact probabilities for generated samples, while allowing low-dimensional latent variables as input. Our experimental results, on various datasets, show that our density estimator produces accurate estimates, while retaining good quality in the generated samples.

Kelian Baert, Shrisha Bharadwaj, Fabien Castan et al.

Feedforward monocular face capture methods seek to reconstruct posed faces from a single image of a person. Current state of the art approaches have the ability to regress parametric 3D face models in real-time across a wide range of identities, lighting conditions and poses by leveraging large image datasets of human faces. These methods however suffer from clear limitations in that the underlying parametric face model only provides a coarse estimation of the face shape, thereby limiting their practical applicability in tasks that require precise 3D reconstruction (aging, face swapping, digital make-up, ...). In this paper, we propose a method for high-precision 3D face capture taking advantage of a collection of unconstrained videos of a subject as prior information. Our proposal builds on a two stage approach. We start with the reconstruction of a detailed 3D face avatar of the person, capturing both precise geometry and appearance from a collection of videos. We then use the encoder from a pre-trained monocular face reconstruction method, substituting its decoder with our personalized model, and proceed with transfer learning on the video collection. Using our pre-estimated image formation model, we obtain a more precise self-supervision objective, enabling improved expression and pose alignment. This results in a trained encoder capable of efficiently regressing pose and expression parameters in real-time from previously unseen images, which combined with our personalized geometry model yields more accurate and high fidelity mesh inference. Through extensive qualitative and quantitative evaluation, we showcase the superiority of our final model as compared to state-of-the-art baselines, and demonstrate its generalization ability to unseen pose, expression and lighting.

Peter Kulits, Haiwen Feng, Weiyang Liu et al.

Inverse graphics -- the task of inverting an image into physical variables that, when rendered, enable reproduction of the observed scene -- is a fundamental challenge in computer vision and graphics. Successfully disentangling an image into its constituent elements, such as the shape, color, and material properties of the objects of the 3D scene that produced it, requires a comprehensive understanding of the environment. This complexity limits the ability of existing carefully engineered approaches to generalize across domains. Inspired by the zero-shot ability of large language models (LLMs) to generalize to novel contexts, we investigate the possibility of leveraging the broad world knowledge encoded in such models to solve inverse-graphics problems. To this end, we propose the Inverse-Graphics Large Language Model (IG-LLM), an inverse-graphics framework centered around an LLM, that autoregressively decodes a visual embedding into a structured, compositional 3D-scene representation. We incorporate a frozen pre-trained visual encoder and a continuous numeric head to enable end-to-end training. Through our investigation, we demonstrate the potential of LLMs to facilitate inverse graphics through next-token prediction, without the application of image-space supervision. Our analysis enables new possibilities for precise spatial reasoning about images that exploit the visual knowledge of LLMs. We release our code and data at https://ig-llm.is.tue.mpg.de/ to ensure the reproducibility of our investigation and to facilitate future research.

George Retsinas, Panagiotis P. Filntisis, Radek Danecek et al.

While existing methods for 3D face reconstruction from in-the-wild images excel at recovering the overall face shape, they commonly miss subtle, extreme, asymmetric, or rarely observed expressions. We improve upon these methods with SMIRK (Spatial Modeling for Image-based Reconstruction of Kinesics), which faithfully reconstructs expressive 3D faces from images. We identify two key limitations in existing methods: shortcomings in their self-supervised training formulation, and a lack of expression diversity in the training images. For training, most methods employ differentiable rendering to compare a predicted face mesh with the input image, along with a plethora of additional loss functions. This differentiable rendering loss not only has to provide supervision to optimize for 3D face geometry, camera, albedo, and lighting, which is an ill-posed optimization problem, but the domain gap between rendering and input image further hinders the learning process. Instead, SMIRK replaces the differentiable rendering with a neural rendering module that, given the rendered predicted mesh geometry, and sparsely sampled pixels of the input image, generates a face image. As the neural rendering gets color information from sampled image pixels, supervising with neural rendering-based reconstruction loss can focus solely on the geometry. Further, it enables us to generate images of the input identity with varying expressions while training. These are then utilized as input to the reconstruction model and used as supervision with ground truth geometry. This effectively augments the training data and enhances the generalization for diverse expressions. Our qualitative, quantitative and particularly our perceptual evaluations demonstrate that SMIRK achieves the new state-of-the art performance on accurate expression reconstruction. Project webpage: https://georgeretsi.github.io/smirk/.

Xi Wang, Nicolas Dufour, Nefeli Andreou et al.

Classifier-Free Guidance (CFG) enhances the quality and condition adherence of text-to-image diffusion models. It operates by combining the conditional and unconditional predictions using a fixed weight. However, recent works vary the weights throughout the diffusion process, reporting superior results but without providing any rationale or analysis. By conducting comprehensive experiments, this paper provides insights into CFG weight schedulers. Our findings suggest that simple, monotonically increasing weight schedulers consistently lead to improved performances, requiring merely a single line of code. In addition, more complex parametrized schedulers can be optimized for further improvement, but do not generalize across different models and tasks.

Haiwen Feng, Zheng Ding, Zhihao Xia et al.

We introduce bounded generation as a generalized task to control video generation to synthesize arbitrary camera and subject motion based only on a given start and end frame. Our objective is to fully leverage the inherent generalization capability of an image-to-video model without additional training or fine-tuning of the original model. This is achieved through the proposed new sampling strategy, which we call Time Reversal Fusion, that fuses the temporally forward and backward denoising paths conditioned on the start and end frame, respectively. The fused path results in a video that smoothly connects the two frames, generating inbetweening of faithful subject motion, novel views of static scenes, and seamless video looping when the two bounding frames are identical. We curate a diverse evaluation dataset of image pairs and compare against the closest existing methods. We find that Time Reversal Fusion outperforms related work on all subtasks, exhibiting the ability to generate complex motions and 3D-consistent views guided by bounded frames. See project page at https://time-reversal.github.io.

Nefeli Andreou, Xi Wang, Victoria Fernández Abrevaya et al.

Our goal is to generate realistic human motion from natural language. Modern methods often face a trade-off between model expressiveness and text-to-motion alignment. Some align text and motion latent spaces but sacrifice expressiveness; others rely on diffusion models producing impressive motions, but lacking semantic meaning in their latent space. This may compromise realism, diversity, and applicability. Here, we address this by combining latent diffusion with a realignment mechanism, producing a novel, semantically structured space that encodes the semantics of language. Leveraging this capability, we introduce the task of textual motion inversion to capture novel motion concepts from a few examples. For motion synthesis, we evaluate LEAD on HumanML3D and KIT-ML and show comparable performance to the state-of-the-art in terms of realism, diversity, and text-motion consistency. Our qualitative analysis and user study reveal that our synthesized motions are sharper, more human-like and comply better with the text compared to modern methods. For motion textual inversion, our method demonstrates improved capacity in capturing out-of-distribution characteristics in comparison to traditional VAEs.

Rick Akkerman, Haiwen Feng, Michael J. Black et al.

Predicting the dynamics of interacting objects is essential for both humans and intelligent systems. However, existing approaches are limited to simplified, toy settings and lack generalizability to complex, real-world environments. Recent advances in generative models have enabled the prediction of state transitions based on interventions, but focus on generating a single future state which neglects the continuous dynamics resulting from the interaction. To address this gap, we propose InterDyn, a novel framework that generates videos of interactive dynamics given an initial frame and a control signal encoding the motion of a driving object or actor. Our key insight is that large video generation models can act as both neural renderers and implicit physics ``simulators'', having learned interactive dynamics from large-scale video data. To effectively harness this capability, we introduce an interactive control mechanism that conditions the video generation process on the motion of the driving entity. Qualitative results demonstrate that InterDyn generates plausible, temporally consistent videos of complex object interactions while generalizing to unseen objects. Quantitative evaluations show that InterDyn outperforms baselines that focus on static state transitions. This work highlights the potential of leveraging video generative models as implicit physics engines. Project page: https://interdyn.is.tue.mpg.de/

Shrisha Bharadwaj, Haiwen Feng, Giorgio Becherini et al.

Manipulating the illumination of a 3D scene within a single image represents a fundamental challenge in computer vision and graphics. This problem has traditionally been addressed using inverse rendering techniques, which involve explicit 3D asset reconstruction and costly ray-tracing simulations. Meanwhile, recent advancements in visual foundation models suggest that a new paradigm could soon be possible -- one that replaces explicit physical models with networks that are trained on large amounts of image and video data. In this paper, we exploit the implicit scene understanding of a video diffusion model, particularly Stable Video Diffusion, to relight a single image. We introduce GenLit, a framework that distills the ability of a graphics engine to perform light manipulation into a video-generation model, enabling users to directly insert and manipulate a point light in the 3D world within a given image and generate results directly as a video sequence. We find that a model fine-tuned on only a small synthetic dataset generalizes to real-world scenes, enabling single-image relighting with plausible and convincing shadows and inter-reflections. Our results highlight the ability of video foundation models to capture rich information about lighting, material, and shape, and our findings indicate that such models, with minimal training, can be used to perform relighting without explicit asset reconstruction or ray-tracing. . Project page: https://genlit.is.tue.mpg.de/.

Pratheba Selvaraju, Victoria Fernandez Abrevaya, Timo Bolkart et al.

Reconstructing 3D face models from a single image is an inherently ill-posed problem, which becomes even more challenging in the presence of occlusions. In addition to fewer available observations, occlusions introduce an extra source of ambiguity where multiple reconstructions can be equally valid. Despite the ubiquity of the problem, very few methods address its multi-hypothesis nature. In this paper we introduce OFER, a novel approach for single-image 3D face reconstruction that can generate plausible, diverse, and expressive 3D faces, even under strong occlusions. Specifically, we train two diffusion models to generate the shape and expression coefficients of a face parametric model, conditioned on the input image. This approach captures the multi-modal nature of the problem, generating a distribution of solutions as output. However, to maintain consistency across diverse expressions, the challenge is to select the best matching shape. To achieve this, we propose a novel ranking mechanism that sorts the outputs of the shape diffusion network based on predicted shape accuracy scores. We evaluate our method using standard benchmarks and introduce CO-545, a new protocol and dataset designed to assess the accuracy of expressive faces under occlusion. Our results show improved performance over occlusion-based methods, while also enabling the generation of diverse expressions for a given image.

Qian li, Victoria Fernàndez Abrevaya, Franck Multon et al.

We present a method for recovering the shape and radiance of a scene consisting of multiple people given solely a few images. Multi-human scenes are complex due to additional occlusion and clutter. For single-human settings, existing approaches using implicit neural representations have achieved impressive results that deliver accurate geometry and appearance. However, it remains challenging to extend these methods for estimating multiple humans from sparse views. We propose a neural implicit reconstruction method that addresses the inherent challenges of this task through the following contributions: First, we propose to use geometry constraints by exploiting pre-computed meshes using a human body model (SMPL). Specifically, we regularize the signed distances using the SMPL mesh and leverage bounding boxes for improved rendering. Second, we propose a ray regularization scheme to minimize rendering inconsistencies, and a saturation regularization for robust optimization in variable illumination. Extensive experiments on both real and synthetic datasets demonstrate the benefits of our approach and show state-of-the-art performance against existing neural reconstruction methods.

Yufeng Zheng, Victoria Fernández Abrevaya, Marcel C. Bühler et al.

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.

Victoria Fernandez Abrevaya, Adnane Boukhayma, Philip H. S. Torr et al.

We present an approach for estimating surface normals from in-the-wild color images of faces. While data-driven strategies have been proposed for single face images, limited available ground truth data makes this problem difficult. To alleviate this issue, we propose a method that can leverage all available image and normal data, whether paired or not, thanks to a novel cross-modal learning architecture. In particular, we enable additional training with single modality data, either color or normal, by using two encoder-decoder networks with a shared latent space. The proposed architecture also enables face details to be transferred between the image and normal domains, given paired data, through skip connections between the image encoder and normal decoder. Core to our approach is a novel module that we call deactivable skip connections, which allows integrating both the auto-encoded and image-to-normal branches within the same architecture that can be trained end-to-end. This allows learning of a rich latent space that can accurately capture the normal information. We compare against state-of-the-art methods and show that our approach can achieve significant improvements, both quantitative and qualitative, with natural face images.

Victoria Fernandez Abrevaya, Adnane Boukhayma, Stefanie Wuhrer et al.

Data-driven generative 3D face models are used to compactly encode facial shape data into meaningful parametric representations. A desirable property of these models is their ability to effectively decouple natural sources of variation, in particular identity and expression. While factorized representations have been proposed for that purpose, they are still limited in the variability they can capture and may present modeling artifacts when applied to tasks such as expression transfer. In this work, we explore a new direction with Generative Adversarial Networks and show that they contribute to better face modeling performances, especially in decoupling natural factors, while also achieving more diverse samples. To train the model we introduce a novel architecture that combines a 3D generator with a 2D discriminator that leverages conventional CNNs, where the two components are bridged by a geometry mapping layer. We further present a training scheme, based on auxiliary classifiers, to explicitly disentangle identity and expression attributes. Through quantitative and qualitative results on standard face datasets, we illustrate the benefits of our model and demonstrate that it outperforms competing state of the art methods in terms of decoupling and diversity.