13 results
(View BibTeX file of all listed publications)
2020
Egocentric Videoconferencing
Elgharib, M., Mendiratta, M., Thies, J., Nie/ssner, M., Seidel, H., Tewari, A., Vladislav Golyanik, , Theobalt, C.
Siggraph Asia, 2020 (article)
We introduce a method for egocentric videoconferencing that enables hands-free video calls, for instance by people wearing smart glasses or other mixed-reality devices. Videoconferencing portrays valuable non-verbal communication and face expression cues, but usually requires a front-facing camera. Using a frontal camera in a hands-free setting when a person is on the move is impractical. Even holding a mobile phone camera in the front of the face while sitting for a long duration is not convenient. To overcome these issues, we propose a low-cost wearable egocentric camera setup that can be integrated into smart glasses. Our goal is to mimic a classical video call, and therefore, we transform the egocentric perspective of this camera into a front facing video. To this end, we employ a conditional generative adversarial neural network that learns a transition from the highly distorted egocentric views to frontal views common in videoconferencing. Our approach learns to transfer expression details directly from the egocentric view without using a complex intermediate parametric expressions model, as it is used by related face reenactment methods. We successfully handle subtle expressions, not easily captured by parametric blendshape-based solutions, e.g., tongue movement, eye movements, eye blinking, strong expressions and depth varying movements. To get control over the rigid head movements in the target view, we condition the generator on synthetic renderings of a moving neutral face. This allows us to synthesis results at different head poses. Our technique produces temporally smooth video-realistic renderings in real-time using a video-to-video translation network in conjunction with a temporal discriminator. We demonstrate the improved capabilities of our technique by comparing against related state-of-the art approaches.
2019
SpoC: Spoofing Camera Fingerprints
Cozzolino, D., Thies, J., Rössler, A., Nießner, M., Verdoliva, L.
arXiv, 2019 (article)
Thanks to the fast progress in synthetic media generation, creating realistic false images has become very easy. Such images can be used to wrap “rich” fake news with enhanced credibility, spawning a new wave of high-impact, high-risk misinformation campaigns. Therefore, there is a fast-growing interest in reliable detectors of manipulated media. The most powerful detectors, to date, rely on the subtle traces left by any device on all images acquired by it. In particular, due to proprietary in-camera processes, like demosaicing or compression, each camera model leaves trademark traces that can be exploited for forensic analyses. The absence or distortion of such traces in the target image is a strong hint of manipulation.
In this paper, we challenge such detectors to gain better insight into their vulnerabilities. This is an important study in order to build better forgery detectors able to face malicious attacks. Our proposal consists of a GAN-based approach that injects camera traces into synthetic images. Given a GANgenerated image, we insert the traces of a specific camera model into it and deceive state-of-the-art detectors into believing the image was acquired by that model. Likewise, we deceive independent detectors of synthetic GAN images into believing the image is real. Experiments prove the effectiveness of the proposed method in a wide array of conditions. Moreover, no prior information on the attacked detectors is needed, but only sample images from the target camera.
Deferred Neural Rendering: Image Synthesis using Neural Textures
Thies, J., Zollhöfer, M., Nießner, M.
ACM Transactions on Graphics 2019 (TOG), 2019 (article)
The modern computer graphics pipeline can synthesize images at remarkable visual quality; however, it requires well-defined, high-quality 3D content as input. In this work, we explore the use of imperfect 3D content, for instance, obtained from photo-metric reconstructions with noisy and incomplete surface geometry, while still aiming to produce photo-realistic (re-)renderings. To address this challenging problem, we introduce Deferred Neural Rendering, a new paradigm for image synthesis that combines the traditional graphics pipeline with learnable components. Specifically, we propose Neural Textures, which are learned feature maps that are trained as part of the scene capture process. Similar to traditional textures, neural textures are stored as maps on top of 3D mesh proxies; however, the high-dimensional feature maps contain significantly more information, which can be interpreted by our new deferred neural rendering pipeline. Both neural textures and deferred neural renderer are trained end-to-end, enabling us to synthesize photo-realistic images even when the original 3D content was imperfect. In contrast to traditional, black-box 2D generative neural networks, our 3D representation gives us explicit control over the generated output, and allows for a wide range of application domains. For instance, we can synthesize temporally-consistent video re-renderings of recorded 3D scenes as our representation is inherently embedded in 3D space. This way, neural textures can be utilized to coherently re-render or manipulate existing video content in both static and dynamic environments at real-time rates. We show the effectiveness of our approach in several experiments on novel view synthesis, scene editing, and facial reenactment, and compare to state-of-the-art approaches that leverage the standard graphics pipeline as well as conventional generative neural networks.
2018
FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces
Rössler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., Nießner, M.
arXiv, 2018 (article)
FaceForensics is a video dataset consisting of more than 500,000 frames containing faces from 1004 videos that can be used to study image or video forgeries. To create these videos we use an automatated version of the state of the art Face2Face approach. All videos are downloaded from Youtube and are cut down to short continuous clips that contain mostly frontal faces. In particular, we offer two versions of our dataset: Source-to-Target: where we reenact over 1000 videos with new facial expressions extracted from other videos, which e.g. can be used to train a classifier to detect fake images or videos. Selfreenactment: where we use Face2Face to reenact the facial expressions of videos with their own facial expressions as input to get pairs of videos, which e.g. can be used to train supervised generative refinement models.
State of the Art on Monocular 3D Face Reconstruction, Tracking, and Applications
Zollhöfer, M., Thies, J., Bradley, D., Garrido, P., Beeler, T., Péerez, P., Stamminger, M., Nießner, M., Theobalt, C.
EG, 2018 (article)
The computer graphics and vision communities have dedicated long standing efforts in building computerized tools for reconstructing, tracking, and analyzing human faces based on visual input. Over the past years rapid progress has been made, which led to novel and powerful algorithms that obtain impressive results even in the very challenging case of reconstruction from a single RGB or RGB-D camera. The range of applications is vast and steadily growing as these technologies are further improving in speed, accuracy, and ease of use. Motivated by this rapid progress, this state-of-the-art report summarizes recent trends in monocular facial performance capture and discusses its applications, which range from performance-based animation to real-time facial reenactment. We focus our discussion on methods where the central task is to recover and track a three dimensional model of the human face using optimization-based reconstruction algorithms. We provide an in-depth overview of the underlying concepts of real-world image formation, and we discuss common assumptions and simplifications that make these algorithms practical. In addition, we extensively cover the priors that are used to better constrain the under-constrained monocular reconstruction problem, and discuss the optimization techniques that are employed to recover dense, photo-geometric 3D face models from monocular 2D data. Finally, we discuss a variety of use cases for the reviewed algorithms in the context of motion capture, facial animation, as well as image and video editing.
Deep Video Portraits
Kim, H., Garrido, P., Tewari, A., Xu, W., Thies, J., Nießner, M., Péerez, P., Richardt, C., Zollhöfer, M., Theobalt, C.
ACM Transactions on Graphics 2018 (TOG), 2018 (article)
We present a novel approach that enables photo-realistic re-animation of portrait videos using only an input video. In contrast to existing approaches that are restricted to manipulations of facial expressions only, we are the first to transfer the full 3D head position, head rotation, face expression, eye gaze, and eye blinking from a source actor to a portrait video of a target actor. The core of our approach is a generative neural network with a novel space-time architecture. The network takes as input synthetic renderings of a parametric face model, based on which it predicts photo-realistic video frames for a given target actor. The realism in this rendering-to-video transfer is achieved by careful adversarial training, and as a result, we can create modified target videos that mimic the behavior of the synthetically-created input. In order to enable source-to-target video re-animation, we render a synthetic target video with the reconstructed head animation parameters from a source video, and feed it into the trained network – thus taking full control of the target. With the ability to freely recombine source and target parameters, we are able to demonstrate a large variety of video rewrite applications without explicitly modeling hair, body or background. For instance, we can reenact the full head using interactive user-controlled editing, and realize high-fidelity visual dubbing. To demonstrate the high quality of our output, we conduct an extensive series of experiments and evaluations, where for instance a user study shows that our video edits are hard to detect.
HeadOn: Real-time Reenactment of Human Portrait Videos
Thies, J., Zollhöfer, M., Stamminger, M., Theobalt, C., Nießner, M.
ACM Transactions on Graphics 2018 (TOG), 2018 (article)
We propose HeadOn, the first real-time source-to-target reenactment approach for complete human portrait videos that enables transfer of torso and head motion, face expression, and eye gaze. Given a short RGB-D video of the target actor, we automatically construct a personalized geometry proxy that embeds a parametric head, eye, and kinematic torso model. A novel real-time reenactment algorithm employs this proxy to photo-realistically map the captured motion from the source actor to the target actor. On top of the coarse geometric proxy, we propose a video-based rendering technique that composites the modified target portrait video via view- and pose-dependent texturing, and creates photo-realistic imagery of the target actor under novel torso and head poses, facial expressions, and gaze directions. To this end, we propose a robust tracking of the face and torso of the source actor. We extensively evaluate our approach and show significant improvements in enabling much greater flexibility in creating realistic reenacted output videos.
ForensicTransfer: Weakly-supervised Domain Adaptation for Forgery Detection
Cozzolino, D., Thies, J., Rössler, A., Riess, C., Nießner, M., Verdoliva, L.
arXiv, 2018 (article)
Distinguishing fakes from real images is becoming increasingly difficult as new sophisticated image manipulation approaches come out by the day. Convolutional neural networks (CNN) show excellent performance in detecting image manipulations when they are trained on a specific forgery method. However, on examples from unseen manipulation approaches, their performance drops significantly. To address this limitation in transferability, we introduce ForensicTransfer. ForensicTransfer tackles two challenges in multimedia forensics. First, we devise a learning-based forensic detector which adapts well to new domains, i.e., novel manipulation methods. Second we handle scenarios where only a handful of fake examples are available during training. To this end, we learn a forensic embedding that can be used to distinguish between real and fake imagery. We are using a new autoencoder-based architecture which enforces activations in different parts of a latent vector for the real and fake classes. Together with the constraint of correct reconstruction this ensures that the latent space keeps all the relevant information about the nature of the image. Therefore, the learned embedding acts as a form of anomaly detector; namely, an image manipulated from an unseen method will be detected as fake provided it maps sufficiently far away from the cluster of real images. Comparing with prior works, ForensicTransfer shows significant improvements in transferability, which we demonstrate in a series of experiments on cutting-edge benchmarks. For instance, on unseen examples, we achieve up to 80-85% in terms of accuracy compared to 50-59%, and with only a handful of seen examples, our performance already reaches around 95%.
FaceVR: Real-Time Gaze-Aware Facial Reenactment in Virtual Reality
Thies, J., Zollhöfer, M., Stamminger, M., Theobalt, C., Nießner, M.
ACM Transactions on Graphics 2018 (TOG), 2018 (article)
We propose FaceVR, a novel image-based method that enables video teleconferencing in VR based on self-reenactment. State-of-the-art face tracking methods in the VR context are focused on the animation of rigged 3d avatars. While they achieve good tracking performance the results look cartoonish and not real. In contrast to these model-based approaches, FaceVR enables VR teleconferencing using an image-based technique that results in nearly photo-realistic outputs. The key component of FaceVR is a robust algorithm to perform real-time facial motion capture of an actor who is wearing a head-mounted display (HMD), as well as a new data-driven approach for eye tracking from monocular videos. Based on reenactment of a prerecorded stereo video of the person without the HMD, FaceVR incorporates photo-realistic re-rendering in real time, thus allowing artificial modifications of face and eye appearances. For instance, we can alter facial expressions or change gaze directions in the prerecorded target video. In a live setup, we apply these newly-introduced algorithmic components.
2016
Marker-free motion correction in weight-bearing cone-beam CT of the knee joint
Berger, M., Müller, K., Aichert, A., Unberath, M., Thies, J., Choi, J., Fahrig, R., Maier, A.
Medical Physics, 43, pages: 1235-1248, 2016, UnivIS-Import:2017-12-18:Pub.2016.tech.IMMD.IMMD5.marker (article)
Purpose: To allow for a purely image-based motion estimation and compensation in weight-bearing cone-beam computed tomography of the knee joint. Methods: Weight-bearing imaging of the knee joint in a standing position poses additional requirements for the image reconstruction algorithm. In contrast to supine scans, patient motion needs to be estimated and compensated. The authors propose a method that is based on 2D/3D registration of left and right femur and tibia segmented from a prior, motion-free reconstruction acquired in supine position. Each segmented bone is first roughly aligned to the motion-corrupted reconstruction of a scan in standing or squatting position. Subsequently, a rigid 2D/3D registration is performed for each bone to each of K projection images, estimating 6x4xK motion parameters. The motion of individual bones is combined into global motion fields using thin-plate-spline extrapolation. These can be incorporated into a motion-compensated reconstruction in the backprojection step. The authors performed visual and quantitative comparisons between a state-of-the-art marker-based (MB) method and two variants of the proposed method using gradient correlation (GC) and normalized gradient information (NGI) as similarity measure for the 2D/3D registration. Results: The authors evaluated their method on four acquisitions under different squatting positions of the same patient. All methods showed substantial improvement in image quality compared to the uncorrected reconstructions. Compared to NGI and MB, the GC method showed increased streaking artifacts due to misregistrations in lateral projection images. NGI and MB showed comparable image quality at the bone regions. Because the markers are attached to the skin, the MB method performed better at the surface of the legs where the authors observed slight streaking of the NGI and GC methods. For a quantitative evaluation, the authors computed the universal quality index (UQI) for all bone regions with respect to the motion-free reconstruction. The authors quantitative evaluation over regions around the bones yielded a mean UQI of 18.4 for no correction, 53.3 and 56.1 for the proposed method using GC and NGI, respectively, and 53.7 for the MB reference approach. In contrast to the authors registration-based corrections, the MB reference method caused slight nonrigid deformations at bone outlines when compared to a motion-free reference scan. Conclusions: The authors showed that their method based on the NGI similarity measure yields reconstruction quality close to the MB reference method. In contrast to the MB method, the proposed method does not require any preparation prior to the examination which will improve the clinical workflow and patient comfort. Further, the authors found that the MB method causes small, nonrigid deformations at the bone outline which indicates that markers may not accurately reflect the internal motion close to the knee joint. Therefore, the authors believe that the proposed method is a promising alternative to MB motion management.
2015
Real-time Expression Transfer for Facial Reenactment
Thies, J., Zollhöfer, M., Nießner, M., Valgaerts, L., Stamminger, M., Theobalt, C.
ACM Transactions on Graphics (TOG), 34(6), ACM, 2015 (article)
We present a method for the real-time transfer of facial expressions from an actor in a source video to an actor in a target video, thus enabling the ad-hoc control of the facial expressions of the target actor. The novelty of our approach lies in the transfer and photo-realistic re-rendering of facial deformations and detail into the target video in a way that the newly-synthesized expressions are virtually indistinguishable from a real video. To achieve this, we accurately capture the facial performances of the source and target subjects in real-time using a commodity RGB-D sensor. For each frame, we jointly fit a parametric model for identity, expression, and skin reflectance to the input color and depth data, and also reconstruct the scene lighting. For expression transfer, we compute the difference between the source and target expressions in parameter space, and modify the target parameters to match the source expressions. A major challenge is the convincing re-rendering of the synthesized target face into the corresponding video stream. This requires a careful consideration of the lighting and shading design, which both must correspond to the real-world environment. We demonstrate our method in a live setup, where we modify a video conference feed such that the facial expressions of a different person (e.g., translator) are matched in real-time.
Real-Time Pixel Luminance Optimization for Dynamic Multi-Projection Mapping
Siegl, C., Colaianni, M., Thies, L., Thies, J., Zollhöfer, M., Izadi, S., Stamminger, M., Frank, B.
ACM Transactions on Graphics (TOG), 34(6), ACM, 2015 (article)
Using projection mapping enables us to bring virtual worlds into shared physical spaces. In this paper, we present a novel, adaptable and real-time projection mapping system, which supports multiple projectors and high quality rendering of dynamic content on surfaces of complex geometrical shape. Our system allows for smooth blending across multiple projectors using a new optimization framework that simulates the diffuse direct light transport of the physical world to continuously adapt the color output of each projector pixel. We present a real-time solution to this optimization problem using off-the-shelf graphics hardware, depth cameras and projectors. Our approach enables us to move projectors, depth camera or objects while maintaining the correct illumination, in realtime, without the need for markers on the object. It also allows for projectors to be removed or dynamically added, and provides compelling results with only commodity hardware.
2014
Interactive Model-based Reconstruction of the Human Head using an RGB-D Sensor
Zollhöfer, M., Thies, J., Colaianni, M., Stamminger, M., Greiner, G.
Computer Animation and Virtual Worlds, 25, pages: 213-222, 2014 (article)
We present a novel method for the interactive markerless reconstruction of human heads using a single commodity RGB-D sensor. Our entire reconstruction pipeline is implemented on the graphics processing unit and allows to obtain high-quality reconstructions of the human head using an interactive and intuitive reconstruction paradigm. The core of our method is a fast graphics processing unit-based nonlinear quasi-Newton solver that allows us to leverage all information of the RGB-D stream and fit a statistical head model to the observations at interactive frame rates. By jointly solving for shape, albedo and illumination parameters, we are able to reconstruct high-quality models including illumination corrected textures. All obtained reconstructions have a common topology and can be directly used as assets for games, films and various virtual reality applications. We show motion retargeting, retexturing and relighting examples. The accuracy of the presented algorithm is evaluated by a comparison against ground truth data.