We present ObPose, an unsupervised object-centric inference and generation model which learns 3D-structured latent representations from RGB-D scenes. Inspired by prior art in 2D representation learning, ObPose considers a factorised latent space, separately encoding object location (where) and appearance (what). ObPose further leverages an object's pose (i.e. location and orientation), defined via a minimum volume principle, as a novel inductive bias for learning the where component. To achieve this, we propose an efficient, voxelised approximation approach to recover the object shape directly from a neural radiance field (NeRF). As a consequence, ObPose models each scene as a composition of NeRFs, richly representing individual objects. To evaluate the quality of the learned representations, ObPose is evaluated quantitatively on the YCB, MultiShapeNet, and CLEVR datatasets for unsupervised scene segmentation, outperforming the current state-of-the-art in 3D scene inference (ObSuRF) by a significant margin. Generative results provide qualitative demonstration that the same ObPose model can both generate novel scenes and flexibly edit the objects in them. These capacities again reflect the quality of the learned latents and the benefits of disentangling the where and what components of a scene. Key design choices made in the ObPose encoder are validated with ablations.

The synchronous brain activity measured via MEG (or EEG) can be interpreted as arising from a collection (possibly large) of current dipoles or sources located throughout the cortex. Estimating the number, location, and orientation of these sources remains a challenging task, one that is significantly compounded by the effects of source correlations and the presence of interference from spontaneous brain activity, sensor noise, and other artifacts. This paper derives an empirical Bayesian method for addressing each of these issues in a principled fashion. The resulting algorithm guarantees descent of a cost function uniquely designed to handle unknown orientations and arbitrary correlations. Robust interference suppression is also easily incorporated.

We propose a Deep Reinforcement Learning (Deep RL) algorithm for solving the online 3D bin packing problem for an arbitrary number of bins and any bin size. The focus is on producing decisions that can be physically implemented by a robotic loading arm, a laboratory prototype used for testing the concept. The problem considered in this paper is novel in two ways. First, unlike the traditional 3D bin packing problem, we assume that the entire set of objects to be packed is not known a priori. Instead, a fixed number of upcoming objects is visible to the loading system, and they must be loaded in the order of arrival. Second, the goal is not to move objects from one point to another via a feasible path, but to find a location and orientation for each object that maximises the overall packing efficiency of the bin(s). Finally, the learnt model is designed to work with problem instances of arbitrary size without retraining. Simulation results show that the RL-based method outperforms state-of-the-art online bin packing heuristics in terms of empirical competitive ratio and volume efficiency.

Most structures and materials have defects, and if the conditions are right, these defects can lead to the initiation and propagation of cracks. Finding out where and with what orientation a surface crack is most likely to initiate is a critical part of analyzing and designing a structure. An important quantity to compute in this type of analysis is the energy release rate, which is the energy available for crack propagation. The energy release rate is compared to the fracture toughness, a material property that describes the energy required for a crack to propagate. Calculating the energy release rate for the infinite potential locations and orientations of a surface crack in a 3-D structure using conventional methods is an exhaustive task because a detailed analysis needs to be performed for every crack location and orientation. A new method developed by researchers at the University of Illinois at Urbana-Champaign can pinpoint the location and direction of a critical crack in a structure with a single analysis.

The synchronous brain activity measured via MEG (or EEG) can be interpreted as arising from a collection (possibly large) of current dipoles or sources located throughout the cortex. Estimating the number, location, and orientation of these sources remains a challenging task, one that is significantly compounded by the effects of source correlations and the presence of interference from spontaneous brain activity, sensor noise, and other artifacts. This paper derives an empirical Bayesian method for addressing each of these issues in a principled fashion. The resulting algorithm guarantees descent of a cost function uniquely designed to handle unknown orientations and arbitrary correlations. Robust interference suppression is also easily incorporated.

The steel teeth on mining excavation equipment like rope shovels and front end loaders are wear items that must be replaced as part of regular maintenance. During normal operation, the connection that affixes a tooth to the shovel or loader bucket occasionally fails, causing tooth detachment. A detached tooth presents a serious hazard if it enters the haulage cycle and makes its way into a crushing unit, where it may become stuck and require the dangerous task of manual removal. Furthermore, wayward teeth cause substantial lost time and production due to jammed crushers and damage to downstream processing equipment. Therefore, it is critical to detect when a shovel tooth goes missing as soon as possible so that preventative action may be taken.

We present a new research task and a dataset to understand human social interactions via computational methods, to ultimately endow machines with the ability to encode and decode a broad channel of social signals humans use. This research direction is essential to make a machine that genuinely communicates with humans, which we call Social Artificial Intelligence. We first formulate the "social signal prediction" problem as a way to model the dynamics of social signals exchanged among interacting individuals in a data-driven way. We then present a new 3D motion capture dataset to explore this problem, where the broad spectrum of social signals (3D body, face, and hand motions) are captured in a triadic social interaction scenario. Baseline approaches to predict speaking status, social formation, and body gestures of interacting individuals are presented in the defined social prediction framework.

Posted by Jason A. Polzin, PhD GM Applications and Workflow, GE Healthcare Global Magnetic Resonance Imaging Here we describe our experience using TensorFlow to train a neural network to identify specific anatomy during a brain magnetic resonance imaging (MRI) exam to help improve speed and consistency. MRI (Figure 1.) is a 3D imaging technique that allows clinicians to visualize structures in the body non-invasively and without ionizing radiation. MRI is a widely used and powerful imaging modality due to its superior contrast between "soft" tissues, e.g. One of the key strengths of MRI is being able to image specific locations in the body at an orientation best suited for the purpose of the exam. This means that the operator must plan these scans carefully to yield the best possible images uniquely oriented for each patient to visualize the specific structures that may be of interest.

Despite the ubiquity of drones nowadays, it seems to be generally accepted that learning how to control them properly is just too much work. Consumer drones are increasingly being stuffed full of obstacle-avoidance systems, based on the (likely accurate) assumption that most human pilots are to some degree incompetent. It's not that humans are entirely to blame, because controlling a drone isn't the most intuitive thing in the world, and to make it easier, roboticists have been coming up with all kinds of creative solutions. There's body control, face control, and even brain control, all of which offer various combinations of convenience and capability. The more capability you want in a drone control system, usually the less convenient it is, in that it requires more processing power or infrastructure or brain probes or whatever. Developing a system that's both easy to use and self-contained is quite a challenge, but roboticists from the University of Pennsylvania, U.S. Army Research Laboratory, and New York University are up to it--with just a pair of lightweight gaze-tracking glasses and a small computing unit, a small drone will fly wherever you look.

In this paper we introduce new types of square-piece jigsaw puzzles, where in addition to the unknown location and orientation of each piece, a piece might also need to be flipped. These puzzles, which are associated with a number of real world problems, are considerably harder, from a computational standpoint. Specifically, we present a novel generalized genetic algorithm (GA)-based solver that can handle puzzle pieces of unknown location and orientation (Type 2 puzzles) and (two-sided) puzzle pieces of unknown location, orientation, and face (Type 4 puzzles). To the best of our knowledge, our solver provides a new state-of-the-art, solving previously attempted puzzles faster and far more accurately, handling puzzle sizes that have never been attempted before, and assembling the newly introduced two-sided puzzles automatically and effectively. This paper also presents, among other results, the most extensive set of experimental results, compiled as of yet, on Type 2 puzzles.