Bernhard Egger

Edith Tretschk, Navami Kairanda, Mallikarjun B R et al.

3D reconstruction of deformable (or non-rigid) scenes from a set of monocular 2D image observations is a long-standing and actively researched area of computer vision and graphics. It is an ill-posed inverse problem, since -- without additional prior assumptions -- it permits infinitely many solutions leading to accurate projection to the input 2D images. Non-rigid reconstruction is a foundational building block for downstream applications like robotics, AR/VR, or visual content creation. The key advantage of using monocular cameras is their omnipresence and availability to the end users as well as their ease of use compared to more sophisticated camera set-ups such as stereo or multi-view systems. This survey focuses on state-of-the-art methods for dense non-rigid 3D reconstruction of various deformable objects and composite scenes from monocular videos or sets of monocular views. It reviews the fundamentals of 3D reconstruction and deformation modeling from 2D image observations. We then start from general methods -- that handle arbitrary scenes and make only a few prior assumptions -- and proceed towards techniques making stronger assumptions about the observed objects and types of deformations (e.g. human faces, bodies, hands, and animals). A significant part of this STAR is also devoted to classification and a high-level comparison of the methods, as well as an overview of the datasets for training and evaluation of the discussed techniques. We conclude by discussing open challenges in the field and the social aspects associated with the usage of the reviewed methods.

Skylar Sutherland, Bernhard Egger, Joshua Tenenbaum

We propose a method for constructing generative models of 3D objects from a single 3D mesh and improving them through unsupervised low-shot learning from 2D images. Our method produces a 3D morphable model that represents shape and albedo in terms of Gaussian processes. Whereas previous approaches have typically built 3D morphable models from multiple high-quality 3D scans through principal component analysis, we build 3D morphable models from a single scan or template. As we demonstrate in the face domain, these models can be used to infer 3D reconstructions from 2D data (inverse graphics) or 3D data (registration). Specifically, we show that our approach can be used to perform face recognition using only a single 3D template (one scan total, not one per person). We extend our model to a preliminary unsupervised learning framework that enables the learning of the distribution of 3D faces using one 3D template and a small number of 2D images. This approach could also provide a model for the origins of face perception in human infants, who appear to start with an innate face template and subsequently develop a flexible system for perceiving the 3D structure of any novel face from experience with only 2D images of a relatively small number of familiar faces.

Maximilian Weiherer, Bernhard Egger

Becoming a (super) hero is almost every kid's dream. During their sheltered childhood, they do whatever it takes to grow up to be one. Work hard, play hard -- all day long. But as they're getting older, distractions are more and more likely to occur. They're getting off track. They start discovering what is feared as simple math. Finally, they end up as a researcher, writing boring, non-impressive papers all day long because they only rely on simple mathematics. No top-tier conferences, no respect, no groupies. Life's over. To finally put an end to this tragedy, we propose a fundamentally new algorithm, dubbed zero2hero, that turns every research paper into a scientific masterpiece. Given a LaTeX document containing ridiculously simple math, based on next-generation large language models, our system automatically over-complicates every single equation so that no one, including yourself, is able to understand what the hell is going on. Future reviewers will be blown away by the complexity of your equations, immediately leading to acceptance. zero2hero gets you back on track, because you deserve to be a hero$^{\text{TM}}$. Code leaked at \url{https://github.com/mweiherer/zero2hero}.

Karthik Shetty, Annette Birkhold, Srikrishna Jaganathan et al.

Objective: A digital twin of a patient can be a valuable tool for enhancing clinical tasks such as workflow automation, patient-specific X-ray dose optimization, markerless tracking, positioning, and navigation assistance in image-guided interventions. However, it is crucial that the patient's surface and internal organs are of high quality for any pose and shape estimates. At present, the majority of statistical shape models (SSMs) are restricted to a small number of organs or bones or do not adequately represent the general population. Method: To address this, we propose a deformable human shape and pose model that combines skin, internal organs, and bones, learned from CT images. By modeling the statistical variations in a pose-normalized space using probabilistic PCA while also preserving joint kinematics, our approach offers a holistic representation of the body that can benefit various medical applications. Results: We assessed our model's performance on a registered dataset, utilizing the unified shape space, and noted an average error of 3.6 mm for bones and 8.8 mm for organs. To further verify our findings, we conducted additional tests on publicly available datasets with multi-part segmentations, which confirmed the effectiveness of our model. Conclusion: This works shows that anatomically parameterized statistical shape models can be created accurately and in a computationally efficient manner. Significance: The proposed approach enables the construction of shape models that can be directly applied to various medical applications, including biomechanics and reconstruction.

Karthik Shetty, Annette Birkhold, Srikrishna Jaganathan et al.

We introduce PLIKS (Pseudo-Linear Inverse Kinematic Solver) for reconstruction of a 3D mesh of the human body from a single 2D image. Current techniques directly regress the shape, pose, and translation of a parametric model from an input image through a non-linear mapping with minimal flexibility to any external influences. We approach the task as a model-in-the-loop optimization problem. PLIKS is built on a linearized formulation of the parametric SMPL model. Using PLIKS, we can analytically reconstruct the human model via 2D pixel-aligned vertices. This enables us with the flexibility to use accurate camera calibration information when available. PLIKS offers an easy way to introduce additional constraints such as shape and translation. We present quantitative evaluations which confirm that PLIKS achieves more accurate reconstruction with greater than 10% improvement compared to other state-of-the-art methods with respect to the standard 3D human pose and shape benchmarks while also obtaining a reconstruction error improvement of 12.9 mm on the newer AGORA dataset.

James A. D. Gardner, Bernhard Egger, William A. P. Smith

Inverse rendering is an ill-posed problem. Previous work has sought to resolve this by focussing on priors for object or scene shape or appearance. In this work, we instead focus on a prior for natural illuminations. Current methods rely on spherical harmonic lighting or other generic representations and, at best, a simplistic prior on the parameters. We propose a conditional neural field representation based on a variational auto-decoder with a SIREN network and, extending Vector Neurons, build equivariance directly into the network. Using this, we develop a rotation-equivariant, high dynamic range (HDR) neural illumination model that is compact and able to express complex, high-frequency features of natural environment maps. Training our model on a curated dataset of 1.6K HDR environment maps of natural scenes, we compare it against traditional representations, demonstrate its applicability for an inverse rendering task and show environment map completion from partial observations. A PyTorch implementation, our dataset and trained models can be found at jadgardner.github.io/RENI.

James A. D. Gardner, Evgenii Kashin, Bernhard Egger et al.

Inverse rendering of outdoor scenes from unconstrained image collections is a challenging task, particularly illumination/albedo ambiguities and occlusion of the illumination environment (shadowing) caused by geometry. However, there are many cues in an image that can aid in the disentanglement of geometry, albedo and shadows. Whilst sky is frequently masked out in state-of-the-art methods, we exploit the fact that any sky pixel provides a direct observation of distant lighting in the corresponding direction and, via a neural illumination prior, a statistical cue to derive the remaining illumination environment. The incorporation of our illumination prior is enabled by a novel `outside-in' method for computing differentiable sky visibility based on a neural directional distance function. This is highly efficient and can be trained in parallel with the neural scene representation, allowing gradients from appearance loss to flow from shadows to influence the estimation of illumination and geometry. Our method estimates high-quality albedo, geometry, illumination and sky visibility, achieving state-of-the-art results on the NeRF-OSR relighting benchmark. Our code and models can be found at https://github.com/JADGardner/neusky

Vanessa Wirth, Anna-Maria Liphardt, Birte Coppers et al.

Despite their potential, markerless hand tracking technologies are not yet applied in practice to the diagnosis or monitoring of the activity in inflammatory musculoskeletal diseases. One reason is that the focus of most methods lies in the reconstruction of coarse, plausible poses, whereas in the clinical context, accurate, interpretable, and reliable results are required. Therefore, we propose ShaRPy, the first RGB-D Shape Reconstruction and hand Pose tracking system, which provides uncertainty estimates of the computed pose, e.g., when a finger is hidden or its estimate is inconsistent with the observations in the input, to guide clinical decision-making. Besides pose, ShaRPy approximates a personalized hand shape, promoting a more realistic and intuitive understanding of its digital twin. Our method requires only a light-weight setup with a single consumer-level RGB-D camera yet it is able to distinguish similar poses with only small joint angle deviations in a metrically accurate space. This is achieved by combining a data-driven dense correspondence predictor with traditional energy minimization. To bridge the gap between interactive visualization and biomedical simulation we leverage a parametric hand model in which we incorporate biomedical constraints and optimize for both, its pose and hand shape. We evaluate ShaRPy on a keypoint detection benchmark and show qualitative results of hand function assessments for activity monitoring of musculoskeletal diseases.

James A. D. Gardner, Bernhard Egger, William A. P. Smith

Inverse rendering is an ill-posed problem. Previous work has sought to resolve this by focussing on priors for object or scene shape or appearance. In this work, we instead focus on a prior for natural illuminations. Current methods rely on spherical harmonic lighting or other generic representations and, at best, a simplistic prior on the parameters. This results in limitations for the inverse setting in terms of the expressivity of the illumination conditions, especially when taking specular reflections into account. We propose a conditional neural field representation based on a variational auto-decoder and a transformer decoder. We extend Vector Neurons to build equivariance directly into our architecture, and leveraging insights from depth estimation through a scale-invariant loss function, we enable the accurate representation of High Dynamic Range (HDR) images. The result is a compact, rotation-equivariant HDR neural illumination model capable of capturing complex, high-frequency features in natural environment maps. Training our model on a curated dataset of 1.6K HDR environment maps of natural scenes, we compare it against traditional representations, demonstrate its applicability for an inverse rendering task and show environment map completion from partial observations. We share our PyTorch implementation, dataset and trained models at https://github.com/JADGardner/ns_reni

Hannah Schieber, Fabian Deuser, Bernhard Egger et al.

Novel view synthesis using neural radiance fields (NeRF) is the state-of-the-art technique for generating high-quality images from novel viewpoints. Existing methods require a priori knowledge about extrinsic and intrinsic camera parameters. This limits their applicability to synthetic scenes, or real-world scenarios with the necessity of a preprocessing step. Current research on the joint optimization of camera parameters and NeRF focuses on refining noisy extrinsic camera parameters and often relies on the preprocessing of intrinsic camera parameters. Further approaches are limited to cover only one single camera intrinsic. To address these limitations, we propose a novel end-to-end trainable approach called NeRFtrinsic Four. We utilize Gaussian Fourier features to estimate extrinsic camera parameters and dynamically predict varying intrinsic camera parameters through the supervision of the projection error. Our approach outperforms existing joint optimization methods on LLFF and BLEFF. In addition to these existing datasets, we introduce a new dataset called iFF with varying intrinsic camera parameters. NeRFtrinsic Four is a step forward in joint optimization NeRF-based view synthesis and enables more realistic and flexible rendering in real-world scenarios with varying camera parameters.

Richard Marcus, Niklas Knoop, Bernhard Egger et al.

Virtual testing is a crucial task to ensure safety in autonomous driving, and sensor simulation is an important task in this domain. Most current LiDAR simulations are very simplistic and are mainly used to perform initial tests, while the majority of insights are gathered on the road. In this paper, we propose a lightweight approach for more realistic LiDAR simulation that learns a real sensor's behavior from test drive data and transforms this to the virtual domain. The central idea is to cast the simulation into an image-to-image translation problem. We train our pix2pix based architecture on two real world data sets, namely the popular KITTI data set and the Audi Autonomous Driving Dataset which provide both, RGB and LiDAR images. We apply this network on synthetic renderings and show that it generalizes sufficiently from real images to simulated images. This strategy enables to skip the sensor-specific, expensive and complex LiDAR physics simulation in our synthetic world and avoids oversimplification and a large domain-gap through the clean synthetic environment.

Maximilian Weiherer, Finn Klein, Bernhard Egger

To date, the comparison of Statistical Shape Models (SSMs) is often solely performance-based, carried out by means of simplistic metrics such as compactness, generalization, or specificity. Any similarities or differences between the actual shape spaces can neither be visualized nor quantified. In this paper, we present a new method to qualitatively compare two linear SSMs in dense correspondence by computing approximate intersection spaces and set-theoretic differences between the (hyper-ellipsoidal) allowable shape domains spanned by the models. To this end, we approximate the distribution of shapes lying in the intersection space using Markov chain Monte Carlo and subsequently apply Principal Component Analysis (PCA) to the posterior samples, eventually yielding a new SSM of the intersection space. We estimate differences between linear SSMs in a similar manner; here, however, the resulting spaces are no longer convex and we do not apply PCA but instead use the posterior samples for visualization. We showcase the proposed algorithm qualitatively by computing and analyzing intersection spaces and differences between publicly available face models, focusing on gender-specific male and female as well as identity and expression models. Our quantitative evaluation based on SSMs built from synthetic and real-world data sets provides detailed evidence that the introduced method is able to recover ground-truth intersection spaces and differences accurately.

Aditya Parikh, Misha Sadeghi, Robert Richer et al.

Depression is characterized by persistent sadness and loss of interest, significantly impairing daily functioning and now a widespread mental disorder. Traditional diagnostic methods rely on subjective assessments, necessitating objective approaches for accurate diagnosis. Our study investigates the use of facial action units (AUs) and emotions as biomarkers for depression. We analyzed facial expressions from video data of participants classified with or without depression. Our methodology involved detailed feature extraction, mean intensity comparisons of key AUs, and the application of time series classification models. Furthermore, we employed Principal Component Analysis (PCA) and various clustering algorithms to explore the variability in emotional expression patterns. Results indicate significant differences in the intensities of AUs associated with sadness and happiness between the groups, highlighting the potential of facial analysis in depression assessment.

Svenja Strobel, Matthias Innmann, Bernhard Egger et al.

LiDAR-captured point clouds are often considered the gold standard in active 3D reconstruction. While their accuracy is exceptional in flat regions, the capturing is susceptible to miss small geometric structures and may fail with dark, absorbent materials. Alternatively, capturing multiple photos of the scene and applying 3D photogrammetry can infer these details as they often represent feature-rich regions. However, the accuracy of LiDAR for featureless regions is rarely reached. Therefore, we suggest combining the strengths of LiDAR and camera-based capture by introducing SurfFill: a Gaussian surfel-based LiDAR completion scheme. We analyze LiDAR capturings and attribute LiDAR beam divergence as a main factor for artifacts, manifesting mostly at thin structures and edges. We use this insight to introduce an ambiguity heuristic for completed scans by evaluating the change in density in the point cloud. This allows us to identify points close to missed areas, which we can then use to grow additional points from to complete the scan. For this point growing, we constrain Gaussian surfel reconstruction [Huang et al. 2024] to focus optimization and densification on these ambiguous areas. Finally, Gaussian primitives of the reconstruction in ambiguous areas are extracted and sampled for points to complete the point cloud. To address the challenges of large-scale reconstruction, we extend this pipeline with a divide-and-conquer scheme for building-sized point cloud completion. We evaluate on the task of LiDAR point cloud completion of synthetic and real-world scenes and find that our method outperforms previous reconstruction methods.

Benjamin Cappell, Andreas Stoll, Williams Chukwudi Umah et al.

Computational models trained on a large amount of natural images are the state-of-the-art to study human vision - usually adult vision. Computational models of infant vision and its further development are gaining more and more attention in the community. In this work we aim at the very beginning of our visual experience - pre- and post-natal retinal waves which suggest to be a pre-training mechanism for the primate visual system at a very early stage of development. We see this approach as an instance of biologically plausible data driven inductive bias through pre-training. We built a computational model that mimics this development mechanism by pre-training different artificial convolutional neural networks with simulated retinal wave images. The resulting features of this biologically plausible pre-training closely match the V1 features of the primate visual system. We show that the performance gain by pre-training with retinal waves is similar to a state-of-the art pre-training pipeline. Our framework contains the retinal wave generator, as well as a training strategy, which can be a first step in a curriculum learning based training diet for various models of development. We release code, data and trained networks to build the basis for future work on visual development and based on a curriculum learning approach including prenatal development to support studies of innate vs. learned properties of the primate visual system. An additional benefit of our pre-trained networks for neuroscience or computer vision applications is the absence of biases inherited from datasets like ImageNet.

Danrui Li, Jiahao Zhang, Bernhard Egger et al.

Assembling objects from parts requires understanding multimodal instructions, linking them to 3D components, and predicting physically plausible 6-DoF motions for each assembly step. Existing datasets focus on simplified scenarios, overlooking shape complexities and assembly trajectories in industrial assemblies. We introduce AssemblyBench, a synthetic dataset of 2,789 industrial objects with multimodal instruction manuals, corresponding 3D part models, and part assembly trajectories. We also propose a transformer-based model, AssemblyDyno, which uses the instructional manual and the 3D shape of each part to jointly predict assembly order and part assembly trajectories. AssemblyDyno outperforms prior works in both assembly pose estimation and trajectory feasibility, where the latter is evaluated by our physics-based simulations.

Maximilian Weiherer, Antonia von Riedheim, Vanessa Brébant et al.

We present a neural parametric 3D breast shape model and, based on this model, introduce a low-cost and accessible 3D surface reconstruction pipeline capable of recovering accurate breast geometry from a monocular RGB video. In contrast to widely used, commercially available yet prohibitively expensive 3D breast scanning solutions and existing low-cost alternatives, our method requires neither specialized hardware nor proprietary software and can be used with any device that is able to record RGB videos. The key building blocks of our pipeline are a state-of-the-art, off-the-shelf Structure-from-motion pipeline, paired with a parametric breast model for robust and metrically correct surface reconstruction. Our model, similarly to the recently proposed implicit Regensburg Breast Shape Model (iRBSM), leverages implicit neural representations to model breast shapes. However, unlike the iRBSM, which employs a single global neural signed distance function (SDF), our approach -- inspired by recent state-of-the-art face models -- decomposes the implicit breast domain into multiple smaller regions, each represented by a local neural SDF anchored at anatomical landmark positions. When incorporated into our surface reconstruction pipeline, the proposed model, dubbed liRBSM (short for localized iRBSM), significantly outperforms the iRBSM in terms of reconstruction quality, yielding more detailed surface reconstruction than its global counterpart. Overall, we find that the introduced pipeline is able to recover high-quality 3D breast geometry within an error margin of less than 2 mm. Our method is fast (requires less than six minutes), fully transparent and open-source, and -- together with the model -- publicly available at https://rbsm.re-mic.de/local-implicit.

Maximilian Hahn, Bernhard Egger

Statistical Shape Models of faces and various body parts are heavily used in medical image analysis, computer vision and visualization. Whilst the field is well explored with many existing tools, all of them aim at experts, which limits their applicability. We demonstrate the first tool that enables the convenient exploration of statistical shape models in the browser, with the capability to manipulate the faces in a targeted manner. This manipulation is performed via a posterior model given partial observations. We release our code and application on GitHub https://github.com/maximilian-hahn/exploreCOSMOS

Thomas Gerig, Andreas Morel-Forster, Clemens Blumer et al.

In this paper, we present a novel open-source pipeline for face registration based on Gaussian processes as well as an application to face image analysis. Non-rigid registration of faces is significant for many applications in computer vision, such as the construction of 3D Morphable face models (3DMMs). Gaussian Process Morphable Models (GPMMs) unify a variety of non-rigid deformation models with B-splines and PCA models as examples. GPMM separate problem specific requirements from the registration algorithm by incorporating domain-specific adaptions as a prior model. The novelties of this paper are the following: (i) We present a strategy and modeling technique for face registration that considers symmetry, multi-scale and spatially-varying details. The registration is applied to neutral faces and facial expressions. (ii) We release an open-source software framework for registration and model-building, demonstrated on the publicly available BU3D-FE database. The released pipeline also contains an implementation of an Analysis-by-Synthesis model adaption of 2D face images, tested on the Multi-PIE and LFW database. This enables the community to reproduce, evaluate and compare the individual steps of registration to model-building and 3D/2D model fitting. (iii) Along with the framework release, we publish a new version of the Basel Face Model (BFM-2017) with an improved age distribution and an additional facial expression model.

Paul Walker, James A. D. Gardner, Andreea Ardelean et al.

Inverse rendering is an ill-posed problem, but priors like illumination priors, can simplify it. Existing work either disregards the spherical and rotation-equivariant nature of illumination environments or does not provide a well-behaved latent space. We propose a rotation-equivariant variational autoencoder that models natural illumination on the sphere without relying on 2D projections. To preserve the SO(2)-equivariance of environment maps, we use a novel Vector Neuron Vision Transformer (VN-ViT) as encoder and a rotation-equivariant conditional neural field as decoder. In the encoder, we reduce the equivariance from SO(3) to SO(2) using a novel SO(2)-equivariant fully connected layer, an extension of Vector Neurons. We show that our SO(2)-equivariant fully connected layer outperforms standard Vector Neurons when used in our SO(2)-equivariant model. Compared to previous methods, our variational autoencoder enables smoother interpolation in latent space and offers a more well-behaved latent space.

Maximilian Weiherer, Bernhard Egger

We propose to use the family of Matérn kernels for implicit surface reconstruction, building upon the recent success of kernel methods for 3D reconstruction of oriented point clouds. As we show from a theoretical and practical perspective, Matérn kernels have some appealing properties which make them particularly well suited for surface reconstruction -- outperforming state-of-the-art methods based on the arc-cosine kernel while being significantly easier to implement, faster to compute, and scalable. Being stationary, we demonstrate that Matérn kernels allow for tunable surface reconstruction in the same way as Fourier feature mappings help coordinate-based MLPs overcome spectral bias. Moreover, we theoretically analyze Matérn kernels' connection to SIREN networks as well as their relation to previously employed arc-cosine kernels. Finally, based on recently introduced Neural Kernel Fields, we present data-dependent Matérn kernels and conclude that especially the Laplace kernel (being part of the Matérn family) is extremely competitive, performing almost on par with state-of-the-art methods in the noise-free case while having a more than five times shorter training time.

Valentin Braeutigam, Matthias Stock, Bernhard Egger

Most currently used object detection methods are learning-based, and can detect objects under varying appearances. Those models require training and a training dataset. We focus on use cases with less data variation, but the requirement of being free of generation of training data and training. Such a setup is for example desired in automatic testing of graphical interfaces during software development, especially for continuous integration testing. In our approach, we use segments from segmentation foundation models and combine them with a simple feature-based classification method. This saves time and cost when changing the object to be searched or its design, as nothing has to be retrained and no dataset has to be created. We evaluate our method on the task of detecting and classifying icons in navigation maps, which is used to simplify and automate the testing of user interfaces in automotive industry. Our methods achieve results almost on par with learning-based object detection methods like YOLO, without the need for training.

Haomiao Ni, Bernhard Egger, Suhas Lohit et al.

Text-conditioned image-to-video generation (TI2V) aims to synthesize a realistic video starting from a given image (e.g., a woman's photo) and a text description (e.g., "a woman is drinking water."). Existing TI2V frameworks often require costly training on video-text datasets and specific model designs for text and image conditioning. In this paper, we propose TI2V-Zero, a zero-shot, tuning-free method that empowers a pretrained text-to-video (T2V) diffusion model to be conditioned on a provided image, enabling TI2V generation without any optimization, fine-tuning, or introducing external modules. Our approach leverages a pretrained T2V diffusion foundation model as the generative prior. To guide video generation with the additional image input, we propose a "repeat-and-slide" strategy that modulates the reverse denoising process, allowing the frozen diffusion model to synthesize a video frame-by-frame starting from the provided image. To ensure temporal continuity, we employ a DDPM inversion strategy to initialize Gaussian noise for each newly synthesized frame and a resampling technique to help preserve visual details. We conduct comprehensive experiments on both domain-specific and open-domain datasets, where TI2V-Zero consistently outperforms a recent open-domain TI2V model. Furthermore, we show that TI2V-Zero can seamlessly extend to other tasks such as video infilling and prediction when provided with more images. Its autoregressive design also supports long video generation.

Andreea Ardelean, Mert Özer, Bernhard Egger

Single-view 3D reconstruction is currently approached from two dominant perspectives: reconstruction of scenes with limited diversity using 3D data supervision or reconstruction of diverse singular objects using large image priors. However, real-world scenarios are far more complex and exceed the capabilities of these methods. We therefore propose a hybrid method following a divide-and-conquer strategy. We first process the scene holistically, extracting depth and semantic information, and then leverage an object-level method for the detailed reconstruction of individual components. By splitting the problem into simpler tasks, our system is able to generalize to various types of scenes without retraining or fine-tuning. We purposely design our pipeline to be highly modular with independent, self-contained modules, to avoid the need for end-to-end training of the whole system. This enables the pipeline to naturally improve as future methods can replace the individual modules. We demonstrate the reconstruction performance of our approach on both synthetic and real-world scenes, comparing favorable against prior works. Project page: https://andreeadogaru.github.io/Gen3DSR

Frederik Riedmann, Bernhard Egger, Tim Rohe

Previous studies have shown that faces are rated as more attractive when they are partially occluded. The cause of this observation remains unclear. One explanation is a mental reconstruction of the occluded face parts which is biased towards a more attractive percept as shown in face-attractiveness rating tasks. We aimed to test for this hypothesis by using a delayed matching-to-sample task, which directly requires mental reconstruction. In two online experiments, we presented observers with unattractive, neutral or attractive synthetic reconstructions of the occluded face parts using a state-of-the-art diffusion-based image generator. Our experiments do not support the initial hypothesis and reveal an unattractiveness bias for occluded faces instead. This suggests that facial attractiveness rating tasks do not prompt reconstructions. Rather, the attractivity bias may arise from global image features, and faces may actually be reconstructed with unattractive properties when mental reconstruction is applied.

Shiyu Li, Hannah Schieber, Niklas Corell et al.

Guidance for assemblable parts is a promising field for augmented reality. Augmented reality assembly guidance requires 6D object poses of target objects in real time. Especially in time-critical medical or industrial settings, continuous and markerless tracking of individual parts is essential to visualize instructions superimposed on or next to the target object parts. In this regard, occlusions by the user's hand or other objects and the complexity of different assembly states complicate robust and real-time markerless multi-object tracking. To address this problem, we present Graph-based Object Tracking (GBOT), a novel graph-based single-view RGB-D tracking approach. The real-time markerless multi-object tracking is initialized via 6D pose estimation and updates the graph-based assembly poses. The tracking through various assembly states is achieved by our novel multi-state assembly graph. We update the multi-state assembly graph by utilizing the relative poses of the individual assembly parts. Linking the individual objects in this graph enables more robust object tracking during the assembly process. For evaluation, we introduce a synthetic dataset of publicly available and 3D printable assembly assets as a benchmark for future work. Quantitative experiments in synthetic data and further qualitative study in real test data show that GBOT can outperform existing work towards enabling context-aware augmented reality assembly guidance. Dataset and code will be made publically available.

Mert Özer, Maximilian Weiherer, Martin Hundhausen et al.

Neural Radiance Fields (NeRFs) quickly evolved as the new de-facto standard for the task of novel view synthesis when trained on a set of RGB images. In this paper, we conduct a comprehensive evaluation of neural scene representations, such as NeRFs, in the context of multi-modal learning. Specifically, we present four different strategies of how to incorporate a second modality, other than RGB, into NeRFs: (1) training from scratch independently on both modalities; (2) pre-training on RGB and fine-tuning on the second modality; (3) adding a second branch; and (4) adding a separate component to predict (color) values of the additional modality. We chose thermal imaging as second modality since it strongly differs from RGB in terms of radiosity, making it challenging to integrate into neural scene representations. For the evaluation of the proposed strategies, we captured a new publicly available multi-view dataset, ThermalMix, consisting of six common objects and about 360 RGB and thermal images in total. We employ cross-modality calibration prior to data capturing, leading to high-quality alignments between RGB and thermal images. Our findings reveal that adding a second branch to NeRF performs best for novel view synthesis on thermal images while also yielding compelling results on RGB. Finally, we also show that our analysis generalizes to other modalities, including near-infrared images and depth maps. Project page: https://mert-o.github.io/ThermalNeRF/.

Maximilian Weiherer, Antonia von Riedheim, Vanessa Brébant et al.

We present the first deep implicit 3D shape model of the female breast, building upon and improving the recently proposed Regensburg Breast Shape Model (RBSM). Compared to its PCA-based predecessor, our model employs implicit neural representations; hence, it can be trained on raw 3D breast scans and eliminates the need for computationally demanding non-rigid registration -- a task that is particularly difficult for feature-less breast shapes. The resulting model, dubbed iRBSM, captures detailed surface geometry including fine structures such as nipples and belly buttons, is highly expressive, and outperforms the RBSM on different surface reconstruction tasks. Finally, leveraging the iRBSM, we present a prototype application to 3D reconstruct breast shapes from just a single image. Model and code publicly available at https://rbsm.re-mic.de/implicit.

Lukas Meyer, Josef Grün, Maximilian Weiherer et al.

We present MS-Splatting -- a multi-spectral 3D Gaussian Splatting (3DGS) framework that is able to generate multi-view consistent novel views from images of multiple, independent cameras with different spectral domains. In contrast to previous approaches, our method does not require cross-modal camera calibration and is versatile enough to model a variety of different spectra, including thermal and near-infra red, without any algorithmic changes. Unlike existing 3DGS-based frameworks that treat each modality separately (by optimizing per-channel spherical harmonics) and therefore fail to exploit the underlying spectral and spatial correlations, our method leverages a novel neural color representation that encodes multi-spectral information into a learned, compact, per-splat feature embedding. A shallow multi-layer perceptron (MLP) then decodes this embedding to obtain spectral color values, enabling joint learning of all bands within a unified representation. Our experiments show that this simple yet effective strategy is able to improve multi-spectral rendering quality, while also leading to improved per-spectra rendering quality over state-of-the-art methods. We demonstrate the effectiveness of this new technique in agricultural applications to render vegetation indices, such as normalized difference vegetation index (NDVI).

Josef Grün, Lukas Meyer, Maximilian Weiherer et al.

We present a study of how to integrate color (RGB) and multi-spectral imagery (red, green, red-edge, and near-infrared) into the 3D Gaussian Splatting (3DGS) framework, a state-of-the-art explicit radiance-field-based method for fast and high-fidelity 3D reconstruction from multi-view images. While 3DGS excels on RGB data, naive per-band optimization of additional spectra yields poor reconstructions due to inconsistently appearing geometry in the spectral domain. This problem is prominent, even though the actual geometry is the same, regardless of spectral modality. To investigate this, we evaluate three strategies: 1) Separate per-band reconstruction with no shared structure. 2) Splitting optimization, in which we first optimize RGB geometry, copy it, and then fit each new band to the model by optimizing both geometry and band representation. 3) Joint, in which the modalities are jointly optimized, optionally with an initial RGB-only phase. We showcase through quantitative metrics and qualitative novel-view renderings on multi-spectral datasets the effectiveness of our dedicated optimized Joint strategy, increasing overall spectral reconstruction as well as enhancing RGB results through spectral cross-talk. We therefore suggest integrating multi-spectral data directly into the spherical harmonics color components to compactly model each Gaussian's multi-spectral reflectance. Moreover, our analysis reveals several key trade-offs in when and how to introduce spectral bands during optimization, offering practical insights for robust multi-modal 3DGS reconstruction.

Maximilian Weiherer, Andreea Dogaru, Shreya Kapoor et al.

As we all know, writing scientific papers together with our beloved colleagues is a truly remarkable experience (partially): endless discussions about the same useless paragraph over and over again, followed by long days and long nights -- both at the same time. What a wonderful ride it is! What a beautiful life we have. But wait, there's one tiny little problem that utterly shatters the peace, turning even renowned scientists into bloodthirsty monsters: author order. The reason is that, contrary to widespread opinion, it's not the font size that matters, but the way things are ordered. Of course, this is a fairly well-known fact among scientists all across the planet (and beyond) and explains clearly why we regularly have to read about yet another escalated paper submission in local police reports. In this paper, we take an important step backwards to tackle this issue by solving the so-called author ordering problem (AOP) once and for all. Specifically, we propose AMOR, a system that replaces silly constructs like co-first or co-middle authorship with a simple yet easy probabilistic approach based on random shuffling of the author list at viewing time. In addition to AOP, we also solve the ambiguous author ordering citation problem} (AAOCP) on the fly. Stop author violence, be human.

Benno Buschmann, Andreea Dogaru, Elmar Eisemann et al.

Learning-based scene representations such as neural radiance fields or light field networks, that rely on fitting a scene model to image observations, commonly encounter challenges in the presence of inconsistencies within the images caused by occlusions, inaccurately estimated camera parameters or effects like lens flare. To address this challenge, we introduce RANdom RAy Consensus (RANRAC), an efficient approach to eliminate the effect of inconsistent data, thereby taking inspiration from classical RANSAC based outlier detection for model fitting. In contrast to the down-weighting of the effect of outliers based on robust loss formulations, our approach reliably detects and excludes inconsistent perspectives, resulting in clean images without floating artifacts. For this purpose, we formulate a fuzzy adaption of the RANSAC paradigm, enabling its application to large scale models. We interpret the minimal number of samples to determine the model parameters as a tunable hyperparameter, investigate the generation of hypotheses with data-driven models, and analyze the validation of hypotheses in noisy environments. We demonstrate the compatibility and potential of our solution for both photo-realistic robust multi-view reconstruction from real-world images based on neural radiance fields and for single-shot reconstruction based on light-field networks. In particular, the results indicate significant improvements compared to state-of-the-art robust methods for novel-view synthesis on both synthetic and captured scenes with various inconsistencies including occlusions, noisy camera pose estimates, and unfocused perspectives. The results further indicate significant improvements for single-shot reconstruction from occluded images. Project Page: https://bennobuschmann.com/ranrac/

Karthik Shetty, Annette Birkhold, Bernhard Egger et al.

We present a novel approach for 3D human pose estimation by employing probabilistic modeling. This approach leverages the advantages of normalizing flows in non-Euclidean geometries to address uncertain poses. Specifically, our method employs normalizing flow tailored to the SO(3) rotational group, incorporating a coupling mechanism based on the Möbius transformation. This enables the framework to accurately represent any distribution on SO(3), effectively addressing issues related to discontinuities. Additionally, we reinterpret the challenge of reconstructing 3D human figures from 2D pixel-aligned inputs as the task of mapping these inputs to a range of probable poses. This perspective acknowledges the intrinsic ambiguity of the task and facilitates a straightforward integration method for multi-view scenarios. The combination of these strategies showcases the effectiveness of probabilistic models in complex scenarios for human pose estimation techniques. Our approach notably surpasses existing methods in the field of pose estimation. We also validate our methodology on human pose estimation from RGB images as well as medical X-Ray datasets.

Valentin Braeutigam, Vanessa Wirth, Ingrid Ullmann et al.

The 3D reconstruction of faces gains wide attention in computer vision and is used in many fields of application, for example, animation, virtual reality, and even forensics. This work is motivated by monitoring patients in sleep laboratories. Due to their unique characteristics, sensors from the radar domain have advantages compared to optical sensors, namely penetration of electrically non-conductive materials and independence of light. These advantages of radar signals unlock new applications and require adaptation of 3D reconstruction frameworks. We propose a novel model-based method for 3D reconstruction from radar images. We generate a dataset of synthetic radar images with a physics-based but non-differentiable radar renderer. This dataset is used to train a CNN-based encoder to estimate the parameters of a 3D morphable face model. Whilst the encoder alone already leads to strong reconstructions of synthetic data, we extend our reconstruction in an Analysis-by-Synthesis fashion to a model-based autoencoder. This is enabled by learning the rendering process in the decoder, which acts as an object-specific differentiable radar renderer. Subsequently, the combination of both network parts is trained to minimize both, the loss of the parameters and the loss of the resulting reconstructed radar image. This leads to the additional benefit, that at test time the parameters can be further optimized by finetuning the autoencoder unsupervised on the image loss. We evaluated our framework on generated synthetic face images as well as on real radar images with 3D ground truth of four individuals.

Leonhard Rist, Pluvio Stephan, Noah Maul et al.

Tomographic imaging reveals internal structures of 3D objects and is crucial for medical diagnoses. Visualizing the morphology and appearance of non-planar sparse anatomical structures that extend over multiple 2D slices in tomographic volumes is inherently difficult but valuable for decision-making and reporting. Hence, various organ-specific unfolding techniques exist to map their densely sampled 3D surfaces to a distortion-minimized 2D representation. However, there is no versatile framework to flatten complex sparse structures including vascular, duct or bone systems. We deploy a neural field to fit the transformation of the anatomy of interest to a 2D overview image. We further propose distortion regularization strategies and combine geometric with intensity-based loss formulations to also display non-annotated and auxiliary targets. In addition to improved versatility, our unfolding technique outperforms mesh-based baselines for sparse structures w.r.t. peak distortion and our regularization scheme yields smoother transformations compared to Jacobian formulations from neural field-based image registration.

Takuma Nishimura, Andreea Dogaru, Martin Oeggerli et al.

Scanning Electron Microscopes (SEMs) are widely renowned for their ability to analyze the surface structures of microscopic objects, offering the capability to capture highly detailed, yet only grayscale, images. To create more expressive and realistic illustrations, these images are typically manually colorized by an artist with the support of image editing software. This task becomes highly laborious when multiple images of a scanned object require colorization. We propose facilitating this process by using the underlying 3D structure of the microscopic scene to propagate the color information to all the captured images, from as little as one colorized view. We explore several scene representation techniques and achieve high-quality colorized novel view synthesis of a SEM scene. In contrast to prior work, there is no manual intervention or labelling involved in obtaining the 3D representation. This enables an artist to color a single or few views of a sequence and automatically retrieve a fully colored scene or video. Project page: https://ronly2460.github.io/ArCSEM

Moritz Drobnitzky, Jonas Friederich, Bernhard Egger et al.

Strong demand for autonomous vehicles and the wide availability of 3D sensors are continuously fueling the proposal of novel methods for 3D object detection. In this paper, we provide a comprehensive survey of recent developments from 2012-2021 in 3D object detection covering the full pipeline from input data, over data representation and feature extraction to the actual detection modules. We introduce fundamental concepts, focus on a broad range of different approaches that have emerged over the past decade, and propose a systematization that provides a practical framework for comparing these approaches with the goal of guiding future development, evaluation and application activities. Specifically, our survey and systematization of 3D object detection models and methods can help researchers and practitioners to get a quick overview of the field by decomposing 3DOD solutions into more manageable pieces.

Shubhaankar Gupta, Thomas P. O'Connell, Bernhard Egger

Pre-training on large-scale databases consisting of natural images and then fine-tuning them to fit the application at hand, or transfer-learning, is a popular strategy in computer vision. However, Kataoka et al., 2020 introduced a technique to eliminate the need for natural images in supervised deep learning by proposing a novel synthetic, formula-based method to generate 2D fractals as training corpus. Using one synthetically generated fractal for each class, they achieved transfer learning results comparable to models pre-trained on natural images. In this project, we take inspiration from their work and build on this idea -- using 3D procedural object renders. Since the image formation process in the natural world is based on its 3D structure, we expect pre-training with 3D mesh renders to provide an implicit bias leading to better generalization capabilities in a transfer learning setting and that invariances to 3D rotation and illumination are easier to be learned based on 3D data. Similar to the previous work, our training corpus will be fully synthetic and derived from simple procedural strategies; we will go beyond classic data augmentation and also vary illumination and pose which are controllable in our setting and study their effect on transfer learning capabilities in context to prior work. In addition, we will compare the 2D fractal and 3D procedural object networks to human and non-human primate brain data to learn more about the 2D vs. 3D nature of biological vision.

Safa C. Medin, Bernhard Egger, Anoop Cherian et al.

Recent advances in generative adversarial networks (GANs) have led to remarkable achievements in face image synthesis. While methods that use style-based GANs can generate strikingly photorealistic face images, it is often difficult to control the characteristics of the generated faces in a meaningful and disentangled way. Prior approaches aim to achieve such semantic control and disentanglement within the latent space of a previously trained GAN. In contrast, we propose a framework that a priori models physical attributes of the face such as 3D shape, albedo, pose, and lighting explicitly, thus providing disentanglement by design. Our method, MOST-GAN, integrates the expressive power and photorealism of style-based GANs with the physical disentanglement and flexibility of nonlinear 3D morphable models, which we couple with a state-of-the-art 2D hair manipulation network. MOST-GAN achieves photorealistic manipulation of portrait images with fully disentangled 3D control over their physical attributes, enabling extreme manipulation of lighting, facial expression, and pose variations up to full profile view.

Bernhard Egger, Skylar Sutherland, Safa C. Medin et al.

3D Morphable Models are a class of generative models commonly used to model faces. They are typically applied to ill-posed problems such as 3D reconstruction from 2D data. Several ambiguities in this problem's image formation process have been studied explicitly. We demonstrate that non-orthogonality of the variation in identity and expression can cause identity-expression ambiguity in 3D Morphable Models, and that in practice expression and identity are far from orthogonal and can explain each other surprisingly well. Whilst previously reported ambiguities only arise in an inverse rendering setting, identity-expression ambiguity emerges in the 3D shape generation process itself. We demonstrate this effect with 3D shapes directly as well as through an inverse rendering task, and use two popular models built from high quality 3D scans as well as a model built from a large collection of 2D images and videos. We explore this issue's implications for inverse rendering and observe that it cannot be resolved by a purely statistical prior on identity and expression deformations.

Maximilian Weiherer, Andreas Eigenberger, Bernhard Egger et al.

We present the Regensburg Breast Shape Model (RBSM) -- a 3D statistical shape model of the female breast built from 110 breast scans acquired in a standing position, and the first publicly available. Together with the model, a fully automated, pairwise surface registration pipeline used to establish dense correspondence among 3D breast scans is introduced. Our method is computationally efficient and requires only four landmarks to guide the registration process. A major challenge when modeling female breasts from surface-only 3D breast scans is the non-separability of breast and thorax. In order to weaken the strong coupling between breast and surrounding areas, we propose to minimize the variance outside the breast region as much as possible. To achieve this goal, a novel concept called breast probability masks (BPMs) is introduced. A BPM assigns probabilities to each point of a 3D breast scan, telling how likely it is that a particular point belongs to the breast area. During registration, we use BPMs to align the template to the target as accurately as possible inside the breast region and only roughly outside. This simple yet effective strategy significantly reduces the unwanted variance outside the breast region, leading to better statistical shape models in which breast shapes are quite well decoupled from the thorax. The RBSM is thus able to produce a variety of different breast shapes as independently as possible from the shape of the thorax. Our systematic experimental evaluation reveals a generalization ability of 0.17 mm and a specificity of 2.8 mm. To underline the expressiveness of the proposed model, we finally demonstrate in two showcase applications how the RBSM can be used for surgical outcome simulation and the prediction of a missing breast from the remaining one. Our model is available at https://www.rbsm.re-mic.de/.

Chunlu Li, Andreas Morel-Forster, Thomas Vetter et al.

In this work, we aim to enhance model-based face reconstruction by avoiding fitting the model to outliers, i.e. regions that cannot be well-expressed by the model such as occluders or make-up. The core challenge for localizing outliers is that they are highly variable and difficult to annotate. To overcome this challenging problem, we introduce a joint Face-autoencoder and outlier segmentation approach (FOCUS).In particular, we exploit the fact that the outliers cannot be fitted well by the face model and hence can be localized well given a high-quality model fitting. The main challenge is that the model fitting and the outlier segmentation are mutually dependent on each other, and need to be inferred jointly. We resolve this chicken-and-egg problem with an EM-type training strategy, where a face autoencoder is trained jointly with an outlier segmentation network. This leads to a synergistic effect, in which the segmentation network prevents the face encoder from fitting to the outliers, enhancing the reconstruction quality. The improved 3D face reconstruction, in turn, enables the segmentation network to better predict the outliers. To resolve the ambiguity between outliers and regions that are difficult to fit, such as eyebrows, we build a statistical prior from synthetic data that measures the systematic bias in model fitting. Experiments on the NoW testset demonstrate that FOCUS achieves SOTA 3D face reconstruction performance among all baselines that are trained without 3D annotation. Moreover, our results on CelebA-HQ and the AR database show that the segmentation network can localize occluders accurately despite being trained without any segmentation annotation.

Karthik Shetty, Annette Birkhold, Norbert Strobel et al.

Many minimally invasive interventional procedures still rely on 2D fluoroscopic imaging. Generating a patient-specific 3D model from these X-ray projection data would allow to improve the procedural workflow, e.g. by providing assistance functions such as automatic positioning. To accomplish this, two things are required. First, a statistical human shape model of the human anatomy and second, a differentiable X-ray renderer. In this work, we propose a differentiable renderer by deriving the distance travelled by a ray inside mesh structures to generate a distance map. To demonstrate its functioning, we use it for simulating X-ray images from human shape models. Then we show its application by solving the inverse problem, namely reconstructing 3D models from real 2D fluoroscopy images of the pelvis, which is an ideal anatomical structure for patient registration. This is accomplished by an iterative optimization strategy using gradient descent. With the majority of the pelvis being in the fluoroscopic field of view, we achieve a mean Hausdorff distance of 30 mm between the reconstructed model and the ground truth segmentation.

Skylar Sutherland, Bernhard Egger, Joshua Tenenbaum

We propose a method for constructing generative models of 3D objects from a single 3D mesh. Our method produces a 3D morphable model that represents shape and albedo in terms of Gaussian processes. We define the shape deformations in physical (3D) space and the albedo deformations as a combination of physical-space and color-space deformations. Whereas previous approaches have typically built 3D morphable models from multiple high-quality 3D scans through principal component analysis, we build 3D morphable models from a single scan or template. As we demonstrate in the face domain, these models can be used to infer 3D reconstructions from 2D data (inverse graphics) or 3D data (registration). Specifically, we show that our approach can be used to perform face recognition using only a single 3D scan (one scan total, not one per person), and further demonstrate how multiple scans can be incorporated to improve performance without requiring dense correspondence. Our approach enables the synthesis of 3D morphable models for 3D object categories where dense correspondence between multiple scans is unavailable. We demonstrate this by constructing additional 3D morphable models for fish and birds and use them to perform simple inverse rendering tasks. We share the code used to generate these models and to perform our inverse rendering and registration experiments.

Akash Srivastava, Yamini Bansal, Yukun Ding et al.

Current autoencoder-based disentangled representation learning methods achieve disentanglement by penalizing the (aggregate) posterior to encourage statistical independence of the latent factors. This approach introduces a trade-off between disentangled representation learning and reconstruction quality since the model does not have enough capacity to learn correlated latent variables that capture detail information present in most image data. To overcome this trade-off, we present a novel multi-stage modeling approach where the disentangled factors are first learned using a penalty-based disentangled representation learning method; then, the low-quality reconstruction is improved with another deep generative model that is trained to model the missing correlated latent variables, adding detail information while maintaining conditioning on the previously learned disentangled factors. Taken together, our multi-stage modelling approach results in a single, coherent probabilistic model that is theoretically justified by the principal of D-separation and can be realized with a variety of model classes including likelihood-based models such as variational autoencoders, implicit models such as generative adversarial networks, and tractable models like normalizing flows or mixtures of Gaussians. We demonstrate that our multi-stage model has higher reconstruction quality than current state-of-the-art methods with equivalent disentanglement performance across multiple standard benchmarks. In addition, we apply the multi-stage model to generate synthetic tabular datasets, showcasing an enhanced performance over benchmark models across a variety of metrics. The interpretability analysis further indicates that the multi-stage model can effectively uncover distinct and meaningful features of variations from which the original distribution can be recovered.

William A. P. Smith, Alassane Seck, Hannah Dee et al.

In this paper, we bring together two divergent strands of research: photometric face capture and statistical 3D face appearance modelling. We propose a novel lightstage capture and processing pipeline for acquiring ear-to-ear, truly intrinsic diffuse and specular albedo maps that fully factor out the effects of illumination, camera and geometry. Using this pipeline, we capture a dataset of 50 scans and combine them with the only existing publicly available albedo dataset (3DRFE) of 23 scans. This allows us to build the first morphable face albedo model. We believe this is the first statistical analysis of the variability of facial specular albedo maps. This model can be used as a plug in replacement for the texture model of the Basel Face Model (BFM) or FLAME and we make the model publicly available. We ensure careful spectral calibration such that our model is built in a linear sRGB space, suitable for inverse rendering of images taken by typical cameras. We demonstrate our model in a state of the art analysis-by-synthesis 3DMM fitting pipeline, are the first to integrate specular map estimation and outperform the BFM in albedo reconstruction.

Bernhard Egger, William A. P. Smith, Ayush Tewari et al.

In this paper, we provide a detailed survey of 3D Morphable Face Models over the 20 years since they were first proposed. The challenges in building and applying these models, namely capture, modeling, image formation, and image analysis, are still active research topics, and we review the state-of-the-art in each of these areas. We also look ahead, identifying unsolved challenges, proposing directions for future research and highlighting the broad range of current and future applications.

Bernhard Egger, Markus D. Schirmer, Florian Dubost et al.

We propose and demonstrate a joint model of anatomical shapes, image features and clinical indicators for statistical shape modeling and medical image analysis. The key idea is to employ a copula model to separate the joint dependency structure from the marginal distributions of variables of interest. This separation provides flexibility on the assumptions made during the modeling process. The proposed method can handle binary, discrete, ordinal and continuous variables. We demonstrate a simple and efficient way to include binary, discrete and ordinal variables into the modeling. We build Bayesian conditional models based on observed partial clinical indicators, features or shape based on Gaussian processes capturing the dependency structure. We apply the proposed method on a stroke dataset to jointly model the shape of the lateral ventricles, the spatial distribution of the white matter hyperintensity associated with periventricular white matter disease, and clinical indicators. The proposed method yields interpretable joint models for data exploration and patient-specific statistical shape models for medical image analysis.

Adam Kortylewski, Bernhard Egger, Andreas Morel-Forster et al.

It is well known that deep learning approaches to face recognition and facial landmark detection suffer from biases in modern training datasets. In this work, we propose to use synthetic face images to reduce the negative effects of dataset biases on these tasks. Using a 3D morphable face model, we generate large amounts of synthetic face images with full control over facial shape and color, pose, illumination, and background. With a series of experiments, we extensively test the effects of priming deep nets by pre-training them with synthetic faces. We observe the following positive effects for face recognition and facial landmark detection tasks: 1) Priming with synthetic face images improves the performance consistently across all benchmarks because it reduces the negative effects of biases in the training data. 2) Traditional approaches for reducing the damage of dataset bias, such as data augmentation and transfer learning, are less effective than training with synthetic faces. 3) Using synthetic data, we can reduce the size of real-world datasets by 75% for face recognition and by 50% for facial landmark detection while maintaining performance. Thus, offering a means to focus the data collection process on less but higher quality data.

Adam Kortylewski, Andreas Schneider, Thomas Gerig et al.

Recent advances in deep learning have significantly increased the performance of face recognition systems. The performance and reliability of these models depend heavily on the amount and quality of the training data. However, the collection of annotated large datasets does not scale well and the control over the quality of the data decreases with the size of the dataset. In this work, we explore how synthetically generated data can be used to decrease the number of real-world images needed for training deep face recognition systems. In particular, we make use of a 3D morphable face model for the generation of images with arbitrary amounts of facial identities and with full control over image variations, such as pose, illumination, and background. In our experiments with an off-the-shelf face recognition software we observe the following phenomena: 1) The amount of real training data needed to train competitive deep face recognition systems can be reduced significantly. 2) Combining large-scale real-world data with synthetic data leads to an increased performance. 3) Models trained only on synthetic data with strong variations in pose, illumination, and background perform very well across different datasets even without dataset adaptation. 4) The real-to-virtual performance gap can be closed when using synthetic data for pre-training, followed by fine-tuning with real-world images. 5) There are no observable negative effects of pre-training with synthetic data. Thus, any face recognition system in our experiments benefits from using synthetic face images. The synthetic data generator, as well as all experiments, are publicly available.