Touch plays a fundamental role in manipulation for humans; however, machine perception of contact and pressure typically requires invasive sensors. Recent research has shown that deep models can estimate hand pressure based on a single RGB image. However, evaluations have been limited to controlled settings since collecting diverse data with ground-truth pressure measurements is difficult. We present a novel approach that enables diverse data to be captured with only an RGB camera and a cooperative participant. Our key insight is that people can be prompted to apply pressure in a certain way, and this prompt can serve as a weak label to supervise models to perform well under varied conditions. We collect a novel dataset with 51 participants making fingertip contact with diverse objects. Our network, PressureVision++, outperforms human annotators and prior work. We also demonstrate an application of PressureVision++ to mixed reality where pressure estimation allows everyday surfaces to be used as arbitrary touch-sensitive interfaces. Code, data, and models are available online .1

Convolutional neural network inference on video input is computationally expensive and requires high memory bandwidth. Recently, DeltaCNN [26] managed to reduce the cost by only processing pixels with significant updates over the previous frame. However, DeltaCNN relies on static camera input. Moving cameras add new challenges in how to fuse newly unveiled image regions with already processed regions efficiently to minimize the update rate - without increasing memory overhead and without knowing the camera extrinsics of future frames. In this work, we propose MotionDeltaCNN, a sparse CNN inference framework that supports moving cameras. We introduce spherical buffers and padded convolutions to enable seamless fusion of newly unveiled regions and previously processed regions – without increasing memory footprint. Our evaluation shows that we outperform DeltaCNN by up to 90% for moving camera videos.

Convolutional neural network inference on video data requires powerful hardware for real-time processing. Given the inherent coherence across consecutive frames, large parts of a video typically change little. By skipping identical image regions and truncating insignificant pixel updates, computational redundancy can in theory be reduced significantly. However, these theoretical savings have been difficult to translate into practice, as sparse updates hamper computational consistency and memory access coherence; which are key for efficiency on real hardware. With DeltaCNN, we present a sparse convolutional neural network framework that enables sparse frame-by-frame updates to accelerate video inference in practice. We provide sparse implementations for all typical CNN layers and propagate sparse feature updates end-to-end – without accumulating errors over time. DeltaCNN is applicable to all convolutional neural networks without retraining. To the best of our knowledge, we are the first to significantly outperform the dense reference, cuDNN, in practical settings, achieving speedups of up to 7x with only marginal differences in accuracy. Our CUDA kernels and PyTorch extensions can be found at https://github.com/facebookresearch/DeltaCNN.

Soft robotic grippers facilitate contact-rich manipulation, including robust grasping of varied objects. Yet the beneficial compliance of a soft gripper also results in significant deformation that can make precision manipulation challenging. We present visual pressure estimation & control (VPEC), a method that infers pressure applied by a soft gripper using an RGB image from an external camera. We provide results for visual pressure inference when a pneumatic gripper and a tendon-actuated gripper make contact with a flat surface. We also show that VPEC enables precision manipulation via closed-loop control of inferred pressure images. In our evaluation, a mobile manipulator (Stretch RE1 from Hello Robot) uses visual servoing to make contact at a desired pressure; follow a spatial pressure trajectory; and grasp small low-profile objects, including a microSD card, a penny, and a pill. Overall, our results show that visual estimates of applied pressure can enable a soft gripper to perform precision manipulation.

Tracking body and hand motions in 3D space is essential for social and self-presence in augmented and virtual environments. Unlike the popular 3D pose estimation setting, the problem is often formulated as egocentric tracking based on embodied perception (e.g., egocentric cameras, handheld sensors). In this article, we propose a new data-driven framework for egocentric body tracking, targeting challenges of omnipresent occlusions in optimization-based methods (e.g., inverse kinematics solvers). We first collect a large-scale motion capture dataset with both body and finger motions using optical markers and inertial sensors. This dataset focuses on social scenarios and captures ground truth poses under self-occlusions and body-hand interactions. We then simulate the occlusion patterns in head-mounted camera views on the captured ground truth using a ray casting algorithm and learn a deep neural network to infer the occluded body parts. Our experiments show that our method is able to generate high-fidelity embodied poses by applying the proposed method to the task of real-time egocentric body tracking, finger motion synthesis, and 3-point inverse kinematics.

Physical contact between hands and objects plays a critical role in human grasps. We show that optimizing the pose of a hand to achieve expected contact with an object can improve hand poses inferred via image-based methods. Given a hand mesh and an object mesh, a deep model trained on ground truth contact data infers desirable contact across the surfaces of the meshes. Then, ContactOpt efficiently optimizes the pose of the hand to achieve desirable contact using a differentiable contact model. Notably, our contact model encourages mesh interpenetration to approximate deformable soft tissue in the hand. In our evaluations, our methods result in grasps that better match ground truth contact, have lower kinematic error, and are significantly preferred by human participants. Code and models are available online1.

Fully immersive experiences in AR/VR depend on re-constructing the full body pose of the user without restricting their motion. In this paper we study the use of body-worn electromagnetic (EM) field-based sensing for the task of 3D human pose reconstruction. To this end, we present a method to estimate SMPL parameters from 6-12 EM sensors. We leverage a customized wearable system consisting of wireless EM sensors measuring time-synchronized 6D poses at 120 Hz. To provide accurate poses even with little user instrumentation, we adopt a recently proposed hybrid framework, learned gradient descent (LGD), to iteratively estimate SMPL pose and shape from our input measurements. This allows us to harness powerful pose priors to cope with the idiosyncrasies of the input data and achieve accurate pose estimates. The proposed method uses AMASS to synthesize virtual EM-sensor data and we show that it generalizes well to a newly captured real dataset consisting of a total of 36 minutes of motion from 5 subjects. We achieve reconstruction errors as low as 31.8 mm and 13.3 degrees, outperforming both pure learning- and pure optimization-based methods. Code and data is available under https://ait.ethz.ch/projects/2021/em-pose.

We present a system for real-time hand-tracking to drive virtual and augmented reality (VR/AR) experiences. Using four fisheye monochrome cameras, our system generates accurate and low-jitter 3D hand motion across a large working volume for a diverse set of users. We achieve this by proposing neural network architectures for detecting hands and estimating hand keypoint locations. Our hand detection network robustly handles a variety of real world environments. The keypoint estimation network leverages tracking history to produce spatially and temporally consistent poses. We design scalable, semi-automated mechanisms to collect a large and diverse set of ground truth data using a combination of manual annotation and automated tracking. Additionally, we introduce a detection-by-tracking method that increases smoothness while reducing the computational cost; the optimized system runs at 60Hz on PC and 30Hz on a mobile processor. Together, these contributions yield a practical system for capturing a user's hands and is the default feature on the Oculus Quest VR headset powering input and social presence.

Optical marker-based motion capture is the dominant way for obtaining high-fidelity human body animation for special effects, movies, and video games. However, motion capture has seen limited application to the human hand due to the difficulty of automatically identifying (or labeling) identical markers on self-similar fingers. We propose a technique that frames the labeling problem as a keypoint regression problem conducive to a solution using convolutional neural networks. We demonstrate robustness of our labeling solution to occlusion, ghost markers, hand shape, and even motions involving two hands or handheld objects. Our technique is equally applicable to sparse or dense marker sets and can run in real-time to support interaction prototyping with high-fidelity hand tracking and hand presence in virtual reality.

We propose a system for real-time six degrees of freedom (6DoF) tracking of a passive stylus that achieves sub-millimeter accuracy, which is suitable for writing or drawing in mixed reality applications. Our system is particularly easy to implement, requiring only a monocular camera, a 3D printed dodecahedron, and hand-glued binary square markers. The accuracy and performance we achieve are due to model-based tracking using a calibrated model and a combination of sparse pose estimation and dense alignment. We demonstrate the system performance in terms of speed and accuracy on a number of synthetic and real datasets, showing that it can be competitive with state-of-the-art multi-camera motion capture systems. We also demonstrate several applications of the technology ranging from 2D and 3D drawing in VR to general object manipulation and board games.

Physically based simulation of rigid body dynamics is commonly done by time-stepping systems forward in time. In this paper, we propose methods to allow time-stepping rigid body systems back-ward in time. Unfortunately, reverse-time integration of rigid bodies involving frictional contact is mathematically ill-posed, and can lack unique solutions. We instead propose time-reversed rigid body integrators that can sample possible solutions when unique ones do not exist. We also discuss challenges related to dissipation-related energy gain, sensitivity to initial conditions, stacking, constraints and articulation, rolling, sliding, skidding, bouncing, high angular velocities, rapid velocity growth from micro-collisions, and other problems encountered when going against the usual flow of time.

We explore the use of space-time cuts to smoothly transition between stochastic mesh animation clips involving numerous deformable mesh groups while subject to physical constraints. These transitions are used to construct Mesh Ensemble Motion Graphs for interactive data-driven animation of high-dimensional mesh animation datasets, such as those arising from expensive physical simulations of deformable objects blowing in the wind. We formulate the transition computation as an integer programming problem, and introduce a novel randomized algorithm to compute transitions subject to geometric nonpenetration constraints. We present examples for several physically based motion datasets, with real-time display and optional interactive control over wind intensity via transitions between wind levels. We discuss challenges and opportunities for future work and practical application.

Animation techniques for controlling passive simulation are commonly based on an optimization paradigm: the user provides goals a priori , and sophisticated numerical methods minimize a cost function that represents these goals. Unfortunately, for multibody systems with discontinuous contact events these optimization problems can be highly nontrivial to solve, and many-hour offline optimizations, unintuitive parameters, and convergence failures can frustrate end-users and limit usage. On the other hand, users are quite adaptable, and systems which provide interactive feedback via an intuitive interface can leverage the user's own abilities to quickly produce interesting animations. However, the online computation necessary for interactivity limits scene complexity in practice. We introduce Many-Worlds Browsing , a method which circumvents these limits by exploiting the speed of multibody simulators to compute numerous example simulations in parallel (offline and online), and allow the user to browse and modify them interactively. We demonstrate intuitive interfaces through which the user can select among the examples and interactively adjust those parts of the scene that do not match his requirements. We show that using a combination of our techniques, unusual and interesting results can be generated for moderately sized scenes with under an hour of user time. Scalability is demonstrated by sampling much larger scenes using modest offline computations.

We extend approaches for skinning characters to the general setting of skinning deformable mesh animations. We provide an automatic algorithm for generating progressive skinning approximations, that is particularly efficient for pseudo-articulated motions. Our contributions include the use of nonparametric mean shift clustering of high-dimensional mesh rotation sequences to automatically identify statistically relevant bones, and robust least squares methods to determine bone transformations, bone-vertex influence sets, and vertex weight values. We use a low-rank data reduction model defined in the undeformed mesh configuration to provide progressive convergence with a fixed number of bones. We show that the resulting skinned animations enable efficient hardware rendering, rest pose editing, and deformable collision detection. Finally, we present numerous examples where skins were automatically generated using a single set of parameter values.