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

Video-based human motion transfer creates video animations of humans following a source motion. Current methods show remarkable results for tightly-clad subjects. However, the lack of temporally consistent handling of plausible clothing dynamics, including fine and high-frequency details, significantly limits the attainable visual quality. We address these limitations for the first time in the literature and present a new framework which performs high-fidelity and temporally-consistent human motion transfer with natural pose-dependent non-rigid deformations, for several types of loose garments. In contrast to the previous techniques, we perform image generation in three subsequent stages, synthesizing human shape, structure, and appearance. Given a monocular RGB video of an actor, we train a stack of recurrent deep neural networks that generate these intermediate representations from 2D poses and their temporal derivatives. Splitting the difficult motion transfer problem into subtasks that are aware of the temporal motion context helps us to synthesize results with plausible dynamics and pose-dependent detail. It also allows artistic control of results by manipulation of individual framework stages. In the experimental results, we significantly outperform the state-of-the-art in terms of video realism. Our code and data will be made publicly available.

High-Fidelity Neural Human Motion Transfer from Monocular Video

We present EgoRenderer, a system for rendering full-body neural avatars of a person captured by a wearable, egocentric fisheye camera that is mounted on a cap or a VR headset. Our system renders photorealistic novel views of the actor and her motion from arbitrary virtual camera locations. Rendering full-body avatars from such egocentric images come with unique challenges due to the top-down view and large distortions. We tackle these challenges by decomposing the rendering process into several steps, including texture synthesis, pose construction, and neural image translation. For texture synthesis, we propose Ego-DPNet, a neural network that infers dense correspondences between the input fisheye images and an underlying parametric body model, and to extract textures from egocentric inputs. In addition, to encode dynamic appearances, our approach also learns an implicit texture stack that captures detailed appearance variation across poses and viewpoints. For correct pose generation, we first estimate body pose from the egocentric view using a parametric model. We then synthesize an external free-viewpoint pose image by projecting the parametric model to the user-specified target viewpoint. We next combine the target pose image and the textures into a combined feature image, which is transformed into the output color image using a neural image translation network. Experimental evaluations show that EgoRenderer is capable of generating realistic free-viewpoint avatars of a person wearing an egocentric camera. Comparisons to several baselines demonstrate the advantages of our approach.

EgoRenderer: Rendering Human Avatars from Egocentric Camera Images

In this section, we introduce the data collection process for our test dataset. All personal data in our test dataset is collected with an IRB approval. In order to estimate accurate egocentric camera poses and further obtain the ground truth human body poses under the egocentric camera perspective, we mount a calibration board on the head, rigidly attach it to the egocentric camera, and estimate the pose of the egocentric camera with a multi-view capturing system, as shown in Fig. 1. Before the data collection process, we first estimate the transformation matrix Mhead2ego between the calibration board and the fisheye camera with hand-eye calibration [12]. We place a second calibration board on the scene in a place where it can be seen by both the egocentric camera and the studio cameras. We then estimate the relative pose Mego2calib between the egocentric camera and the external calibration board, the relative pose between the studio cameras and the external calibration board Mext2calib, and the relative pose between the studio cameras and the head-mounted calibration board Mext2head. We can obtain the transformation matrix Mhead2ego with:

Scene-aware Egocentric 3D Human Pose Estimation **Supplementary Material**

In the talk, we give a brief overview of the research done in the Computer Graphics Group and the GraphicsOptics-Vision Group of the Max-Planck-Institut für Informatik in the area of 3D Image Analysis and Synthesis. In this context, we address the whole pipeline ranging from the acquisition of computational scene models, over the algorithmic processing of these scene descriptions, to their photo-realistic rendition in the computer. This paper illustrates the questions that we are trying to answer by means of one of our research projects, video-based rendering. We have developed a model-based system to acquire, reconstruct and render free-viewpoint videos of human actors that nicely illustrates the concept of 3D Image Analysis and Synthesis.

3D Image Analysis and Synthesis at MPI Informatik.

Evaluation. As shown in Fig. 1, the output mesh by some TSDF-based (truncated signed distance function) methods [7, 13] is double-layered while the GT mesh is single-layered. Thus, it is not fair if using the original double-layered mesh to perform evaluations directly. Thus, for the double-layered mesh, we remove the back layer according to the visibility of triangle faces from cameras. In other words, if a triangle face is not visible to all cameras, this face will be removed. Another situation is that some areas in the predicted mesh by some methods, such as [7, 15], are out of the scope of the ground truth (GT) mesh as illustrated in the main paper. These areas in the predicted mesh should also be removed to guarantee fair comparisons. To address these issues, we remove faces in the areas in a (predicted) mesh that are not observed in its corresponding GT mesh or not visible to all cameras. Specifically, we clean the mesh following the steps below: (1) we first remove the faces that cannot be observed by any input view; (2) then we filter faces at empty regions of GT mesh by 2D masks, which are rendered from GT mesh for each input view. For each mask pixel, if there is no projection from the GT mesh for its incompleteness, the corresponding value is false, otherwise the value is true.

Christian Theobalt

Papers

High-Fidelity Neural Human Motion Transfer from Monocular Video

EgoRenderer: Rendering Human Avatars from Egocentric Camera Images

Scene-aware Egocentric 3D Human Pose Estimation Supplementary Material

3D Image Analysis and Synthesis at MPI Informatik.

Supplementary of NeuRIS