We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

Machine Learning is the study of methods for programming computers to learn. Computers are applied to a wide range of tasks, and for most of these it is relatively easy for programmers to design and implement the necessary software. However, there are many tasks for which this is difficult or impossible. These can be divided into four general categories. First, there are problems for which there exist no human experts. For example, in modern automated manufacturing facilities, there is a need to predict machine failures before they occur by analyzing sensor readings. Because the machines are new, there are no human experts who can be interviewed by a programmer to provide the knowledge necessary to build a computer system. A machine learning system can study recorded data and subsequent machine failures and learn prediction rules. Second, there are problems where human experts exist, but where they are unable to explain their expertise. This is the case in many perceptual tasks, such as speech recognition, hand-writing recognition, and natural language understanding. Virtually all humans exhibit expert-level abilities on these tasks, but none of them can describe the detailed steps that they follow as they perform them. Fortunately, humans can provide machines with examples of the inputs and correct outputs for these tasks, so machine learning algorithms can learn to map the inputs to the outputs. Third, there are problems where phenomena are changing rapidly. In finance, for example, people would like to predict the future behavior of the stock market, of consumer purchases, or of exchange rates. These behaviors change frequently, so that even if a programmer could construct a good predictive computer program, it would need to be rewritten frequently. A learning program can relieve the programmer of this burden by constantly modifying and tuning a set of learned prediction rules. Fourth, there are applications that need to be customized for each computer user separately. Consider, for example, a program to filter unwanted electronic mail messages. Different users will need different filters. It is unreasonable to expect each user to program his or her own rules, and it is infeasible to provide every user with a software engineer to keep the rules up-to-date. A machine learning system can learn which mail messages the user rejects and maintain the filtering rules automatically. Machine learning addresses many of the same research questions as the fields of statistics, data mining, and psychology, but with differences of emphasis. Statistics focuses on understanding the phenomena that have generated the data, often with the goal of testing different hypotheses about those phenomena. Data mining seeks to find patterns in the data that are understandable by people. Psychological studies of human learning aspire to understand the mechanisms underlying the various learning behaviors exhibited by people (concept learning, skill acquisition, strategy change, etc.).

Machine learning

We present a system for accurate real-time mapping of complex and arbitrary indoor scenes in variable lighting conditions, using only a moving low-cost depth camera and commodity graphics hardware. We fuse all of the depth data streamed from a Kinect sensor into a single global implicit surface model of the observed scene in real-time. The current sensor pose is simultaneously obtained by tracking the live depth frame relative to the global model using a coarse-to-fine iterative closest point (ICP) algorithm, which uses all of the observed depth data available. We demonstrate the advantages of tracking against the growing full surface model compared with frame-to-frame tracking, obtaining tracking and mapping results in constant time within room sized scenes with limited drift and high accuracy. We also show both qualitative and quantitative results relating to various aspects of our tracking and mapping system. Modelling of natural scenes, in real-time with only commodity sensor and GPU hardware, promises an exciting step forward in augmented reality (AR), in particular, it allows dense surfaces to be reconstructed in real-time, with a level of detail and robustness beyond any solution yet presented using passive computer vision.

/pdf/kinectfusion-real-time-dense-surface-mapping-and-tracking-pl8e8pbmre.pdf

KinectFusion: Real-time dense surface mapping and tracking

Humans perceive the three-dimensional structure of the world with apparent ease. However, despite all of the recent advances in computer vision research, the dream of having a computer interpret an image at the same level as a two-year old remains elusive. Why is computer vision such a challenging problem and what is the current state of the art? Computer Vision: Algorithms and Applications explores the variety of techniques commonly used to analyze and interpret images. It also describes challenging real-world applications where vision is being successfully used, both for specialized applications such as medical imaging, and for fun, consumer-level tasks such as image editing and stitching, which students can apply to their own personal photos and videos. More than just a source of recipes, this exceptionally authoritative and comprehensive textbook/reference also takes a scientific approach to basic vision problems, formulating physical models of the imaging process before inverting them to produce descriptions of a scene. These problems are also analyzed using statistical models and solved using rigorous engineering techniques Topics and features: structured to support active curricula and project-oriented courses, with tips in the Introduction for using the book in a variety of customized courses; presents exercises at the end of each chapter with a heavy emphasis on testing algorithms and containing numerous suggestions for small mid-term projects; provides additional material and more detailed mathematical topics in the Appendices, which cover linear algebra, numerical techniques, and Bayesian estimation theory; suggests additional reading at the end of each chapter, including the latest research in each sub-field, in addition to a full Bibliography at the end of the book; supplies supplementary course material for students at the associated website, http://szeliski.org/Book/. Suitable for an upper-level undergraduate or graduate-level course in computer science or engineering, this textbook focuses on basic techniques that work under real-world conditions and encourages students to push their creative boundaries. Its design and exposition also make it eminently suitable as a unique reference to the fundamental techniques and current research literature in computer vision.

/pdf/computer-vision-algorithms-and-applications-25dn6wu83j.pdf

Computer Vision: Algorithms and Applications

“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

State-of-the-art marker-less performance capture algorithms reconstruct detailed human skeletal motion and space-time coherent surface geometry. Despite being a big improvement over marker-based motion capture methods, they are still rarely applied in practical VFX productions as they require ten or more cameras and a studio with controlled lighting or a green screen background. If one was able to capture performances directly on a general set using only the primary stereo camera used for principal photography, many possibilities would open up in virtual production and previsualization, the creation of virtual actors, and video editing during post-production. We describe a new algorithm which works towards this goal. It is able to track skeletal motion and detailed surface geometry of one or more actors from footage recorded with a stereo rig that is allowed to move. It succeeds in general sets with uncontrolled background and uncontrolled illumination, and scenes in which actors strike non-frontal poses. It is one of the first performance capture methods to exploit detailed BRDF information and scene illumination for accurate pose tracking and surface refinement in general scenes. It also relies on a new foreground segmentation approach that combines appearance, stereo, and pose tracking results to segment out actors from the background. Appearance, segmentation, and motion cues are combined in a new pose optimization framework that is robust under uncontrolled lighting, uncontrolled background and very sparse camera views.

On-set performance capture of multiple actors with a stereo camera

We present a dense 3D correspondence finding method that enables spatio-temporally coherent reconstruction of surface animations from multi-view video data. Given as input a sequence of shape-from-silhouette volumes of a moving subject that were reconstructed for each time frame individually, our method establishes dense surface correspondences between subsequent shapes independently of surface discretization. This is achieved in two steps: first, we obtain sparse correspondences from robust optical features between adjacent frames. Second, we generate dense correspondences which serve as map between respective surfaces. By applying this procedure subsequently to all pairs of time steps we can trivially align one shape with all others. Thus, the original input can be reconstructed as a sequence of meshes with constant connectivity and small tangential distortion. We exemplify the performance and accuracy of our method using several synthetic and captured real-world sequences.

/pdf/dense-correspondence-finding-for-parametrization-free-2qwhhvbuwk.pdf

Dense correspondence finding for parametrization-free animation reconstruction from video

Implicit functions represented as deep learning approximations are powerful for reconstructing 3D surfaces. However, they can only produce static surfaces that are not controllable, which provides limited ability to modify the resulting model by editing its pose or shape parameters. Nevertheless, such features are essential in building flexible models for both computer graphics and computer vision. In this work, we present methodology that combines detail-rich implicit functions and parametric representations in order to reconstruct 3D models of people that remain controllable and accurate even in the presence of clothing. Given sparse 3D point clouds sampled on the surface of a dressed person, we use an Implicit Part Network (IP-Net) to jointly predict the outer 3D surface of the dressed person, the inner body surface, and the semantic correspondences to a parametric body model. We subsequently use correspondences to fit the body model to our inner surface and then non-rigidly deform it (under a parametric body + displacement model) to the outer surface in order to capture garment, face and hair detail. In quantitative and qualitative experiments with both full body data and hand scans we show that the proposed methodology generalizes, and is effective even given incomplete point clouds collected from single-view depth images. Our models and code will be publicly released (http://virtualhumans.mpi-inf.mpg.de/ipnet).

Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction

We present Non-Rigid Neural Radiance Fields (NR-NeRF), a reconstruction and novel view synthesis approach for general non-rigid dynamic scenes. Our approach takes RGB images of a dynamic scene as input (e.g., from a monocular video recording), and creates a high-quality space-time geometry and appearance representation. We show that a single handheld consumer-grade camera is sufficient to synthesize sophisticated renderings of a dynamic scene from novel virtual camera views, e.g. a `bullet-time' video effect. NR-NeRF disentangles the dynamic scene into a canonical volume and its deformation. Scene deformation is implemented as ray bending, where straight rays are deformed non-rigidly. We also propose a novel rigidity network to better constrain rigid regions of the scene, leading to more stable results. The ray bending and rigidity network are trained without explicit supervision. Our formulation enables dense correspondence estimation across views and time, and compelling video editing applications such as motion exaggeration. Our code will be open sourced.

Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video.

Intrinsic video decomposition refers to the fundamentally ambiguous task of separating a video stream into its constituent layers, in particular reflectance and shading layers. Such a decomposition is the basis for a variety of video manipulation applications, such as realistic recoloring or retexturing of objects. We present a novel variational approach to tackle this underconstrained inverse problem at real-time frame rates, which enables on-line processing of live video footage. The problem of finding the intrinsic decomposition is formulated as a mixed variational e2-ep-optimization problem based on an objective function that is specifically tailored for fast optimization. To this end, we propose a novel combination of sophisticated local spatial and global spatio-temporal priors resulting in temporally coherent decompositions at real-time frame rates without the need for explicit correspondence search. We tackle the resulting high-dimensional, non-convex optimization problem via a novel data-parallel iteratively reweighted least squares solver that runs on commodity graphics hardware. Real-time performance is obtained by combining a local-global solution strategy with hierarchical coarse-to-fine optimization. Compelling real-time augmented reality applications, such as recoloring, material editing and retexturing, are demonstrated in a live setup. Our qualitative and quantitative evaluation shows that we obtain high-quality real-time decompositions even for challenging sequences. Our method is able to outperform state-of-the-art approaches in terms of runtime and result quality -- even without user guidance such as scribbles.

Christian Theobalt

Papers

On-set performance capture of multiple actors with a stereo camera

Dense correspondence finding for parametrization-free animation reconstruction from video

Combining Implicit Function Learning and Parametric Models for 3D Human Reconstruction

Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video.

Live intrinsic video