Modeling a human avatar that can plausibly deform to articulations is an active area of research. We present PocoLoco - the first template-free, point-based, pose-conditioned generative model for 3D humans in loose clothing. We motivate our work by noting that most methods require a parametric model of the human body to ground pose-dependent deformations. Consequently, they are restricted to modeling clothing that is topologically similar to the naked body and do not extend well to loose clothing. The few methods that attempt to model loose clothing typically require either canonicalization or a UV-parameterization and need to address the challenging problem of explicitly estimating correspondences for the deforming clothes. In this work, we formulate avatar clothing deformation as a conditional point-cloud generation task within the denoising diffusion framework. Crucially, our framework operates directly on unordered point clouds, eliminating the need for a parametric model or a clothing template. This also enables a variety of practical applications, such as point-cloud completion and pose-based editing - important features for virtual human animation. As current datasets for human avatars in loose clothing are far too small for training diffusion models, we release a dataset of two subjects performing various poses in loose clothing with a total of 75K point clouds. By contributing towards tackling the challenging task of effectively modeling loose clothing and expanding the available data for training these models, we aim to set the stage for further innovation in digital humans. The source code is available at https://github.com/sidsunny/pocoloco.

The advances in monocular 3D human pose estimation are dominated by supervised techniques that require large-scale 2D/3D pose annotations. Such methods often behave erratically in the absence of any provision to discard unfamiliar out-of-distribution data. To this end, we cast the 3D human pose learning as an unsupervised domain adaptation problem. We introduce MRP-Net11Project page: https://sites.google.com/view/mrp-net that constitutes a common deep network backbone with two output heads subscribing to two diverse configurations; a) model-free Joint localization and b) model-based parametric regression. Such a design allows us to derive suitable measures to quantify prediction uncertainty at both pose and Joint level granularity. While supervising only on labeled synthetic samples, the adaptation process aims to minimize the uncertainty for the unlabeled target images while maximizing the same for an extreme out-of-distribution dataset (backgrounds). Alongside synthetic-to-real 3D pose adaptation, the Joint-uncertainties allow expanding the adaptation to work on in-the-wild images even in the presence of occlusion and truncation scenarios. We present a comprehensive evaluation of the proposed approach and demonstrate state-of-the-art performance on benchmark datasets.

Available 3D human pose estimation approaches leverage different forms of strong (2D/3D pose) or weak (multi-view or depth) paired supervision. Barring synthetic or in-studio domains, acquiring such supervision for each new target environment is highly inconvenient. To this end, we cast 3D pose learning as a self-supervised adaptation problem that aims to transfer the task knowledge from a labeled source domain to a completely unpaired target. We propose to infer image-to-pose via two explicit mappings viz. image-to-latent and latent-to-pose where the latter is a pre-learned decoder obtained from a prior-enforcing generative adversarial auto-encoder. Next, we introduce relation distillation as a means to align the unpaired cross-modal samples i.e., the unpaired target videos and unpaired 3D pose sequences. To this end, we propose a new set of non-local relations in order to characterize long-range latent pose interactions, unlike general contrastive relations where positive couplings are limited to a local neighborhood structure. Further, we provide an objective way to quantify non-localness in order to select the most effective relation set. We evaluate different self-adaptation settings and demonstrate state-of-the-art 3D human pose estimation performance on standard benchmarks.

Estimation of 3D human pose from monocular image has gained considerable attention, as a key step to several human-centric applications. However, generalizability of human pose estimation models developed using supervision on large-scale in-studio datasets remains questionable, as these models often perform unsatisfactorily on unseen in-the-wild environments. Though weakly-supervised models have been proposed to address this shortcoming, performance of such models relies on availability of paired supervision on some related task, such as 2D pose or multi-view image pairs. In contrast, we propose a novel kinematic-structure-preserved unsupervised 3D pose estimation framework, which is not restrained by any paired or unpaired weak supervisions. Our pose estimation framework relies on a minimal set of prior knowledge that defines the underlying kinematic 3D structure, such as skeletal joint connectivity information with bone-length ratios in a fixed canonical scale. The proposed model employs three consecutive differentiable transformations namely forward-kinematics, camera-projection and spatial-map transformation. This design not only acts as a suitable bottleneck stimulating effective pose disentanglement, but also yields interpretable latent pose representations avoiding training of an explicit latent embedding to pose mapper. Furthermore, devoid of unstable adversarial setup, we re-utilize the decoder to formalize an energy-based loss, which enables us to learn from in-the-wild videos, beyond laboratory settings. Comprehensive experiments demonstrate our state-of-the-art unsupervised and weakly-supervised pose estimation performance on both Human3.6M and MPI-INF-3DHP datasets. Qualitative results on unseen environments further establish our superior generalization ability.

Camera captured human pose is an outcome of several sources of variation. Performance of supervised 3D pose estimation approaches comes at the cost of dispensing with variations, such as shape and appearance, that may be useful for solving other related tasks. As a result, the learned model not only inculcates task-bias but also dataset-bias because of its strong reliance on the annotated samples, which also holds true for weakly-supervised models. Acknowledging this, we propose a self-supervised learning framework to disentangle such variations from unlabeled video frames. We leverage the prior knowledge on human skeleton and poses in the form of a single part-based 2D puppet model, human pose articulation constraints, and a set of unpaired 3D poses. Our differentiable formalization, bridging the representation gap between the 3D pose and spatial part maps, not only facilitates discovery of interpretable pose disentanglement, but also allows us to operate on videos with diverse camera movements. Qualitative results on unseen in-the-wild datasets establish our superior generalization across multiple tasks beyond the primary tasks of 3D pose estimation and part segmentation. Furthermore, we demonstrate state-of-the-art weakly-supervised 3D pose estimation performance on both Human3.6M and MPI-INF-3DHP datasets.

Apache Hive is an open-source relational database system for analytic big-data workloads. In this paper we describe the key innovations on the journey from batch tool to fully fledged enterprise data warehousing system. We present a hybrid architecture that combines traditional MPP techniques with more recent big data and cloud concepts to achieve the scale and performance required by today's analytic applications. We explore the system by detailing enhancements along four main axis: Transactions, optimizer, runtime, and federation. We then provide experimental results to demonstrate the performance of the system for typical workloads and conclude with a look at the community roadmap.

The broad success of Hadoop has led to a fast-evolving and diverse ecosystem of application engines that are building upon the YARN resource management layer. The open-source implementation of MapReduce is being slowly replaced by a collection of engines dedicated to specific verticals. This has led to growing fragmentation and repeated efforts with each new vertical engine re-implementing fundamental features (e.g. fault-tolerance, security, stragglers mitigation, etc.) from scratch.

This paper is concerned with the design and optimization of continuous-time active-RC and gm-C filters. We demonstrate how we can maximize the filters' dynamic range (DR) for a given voltage swing, area and power consumption. Using closed-form symbolic expressions, the optimization problems are formulated as geometric programs (GPs) and mixed-integer GPs (MIGPs) that can be quickly solved to find the globally-optimal solution. The techniques developed in this paper are applied to the design and optimization of an active-RC filter with 1 MHz bandwidth, and a gm-C filter with 30 MHz bandwidth, both of which implement a 5thorder elliptic transfer function. The proposed design flow is general, and can be extended to any filter topology or any set of specifications in order to find a low-noise, linear, low-area, and low-power circuit solution.

The initial design of Apache Hadoop [1] was tightly focused on running massive, MapReduce jobs to process a web crawl. For increasingly diverse companies, Hadoop has become the data and computational agorá---the de facto place where data and computational resources are shared and accessed. This broad adoption and ubiquitous usage has stretched the initial design well beyond its intended target, exposing two key shortcomings: 1) tight coupling of a specific programming model with the resource management infrastructure, forcing developers to abuse the MapReduce programming model, and 2) centralized handling of jobs' control flow, which resulted in endless scalability concerns for the scheduler.

This paper presents the design and optimization of a nested-Miller compensated, three-stage operational transconductance amplifier (OTA) in 90-nm CMOS that is used in a switched-capacitor (SC) gain stage clocked at 200 MHz. Existing design methods for three-stage OTAs lead to sub-optimal solutions because they decouple inter-related metrics like noise and settling performance. In our approach, the problem of finding an optimal design with the best total integrated noise and settling time has been cohesively solved by formulating a nonlinear constrained optimization program. Equality, inequality, and semi-infinite constraints are formed using closed form symbolic expressions obtained by a closed loop analysis of the SC gain stage and the optimization program is solved by using the interior-point algorithm. Simulation results show that the amplifier achieves a ± 0.1% dynamic error settling time of 2.5 ns with a total integrated noise of 240 μVrms, while consuming 5.2 mW from a 1-V power supply.

This paper presents a full duplex radio design using signal inversion and adaptive cancellation. Signal inversion uses a simple design based on a balanced/unbalanced (Balun) transformer. This new design, unlike prior work, supports wideband and high power systems. In theory, this new design has no limitation on bandwidth or power. In practice, we find that the signal inversion technique alone can cancel at least 45dB across a 40MHz bandwidth. Further, combining signal inversion cancellation with cancellation in the digital domain can reduce self-interference by up to 73dB for a 10MHz OFDM signal. This paper also presents a full duplex medium access control (MAC) design and evaluates it using a testbed of 5 prototype full duplex nodes. Full duplex reduces packet losses due to hidden terminals by up to 88%. Full duplex also mitigates unfair channel allocation in AP-based networks, increasing fairness from 0.85 to 0.98 while improving downlink throughput by 110% and uplink throughput by 15%. These experimental results show that a re- design of the wireless network stack to exploit full duplex capability can result in significant improvements in network performance.

This poster presents the design of single channel full-duplex wireless radios. The design uses a combination of RF and baseband techniques to achieve full-duplexing with minimal effect on link reliability. The poster shows two designs with different RF cancellation techniques, Antenna Cancellation and Signal Inversion Cancellation. It also discusses potential MAC and network gains with full-duplexing. It suggests ways in which a full-duplex system can solve some important problems with existing wireless systems including hidden terminals, loss of throughput due to congestion, and large end-to-end delays.

This paper describes a workflow manager developed and deployed at Yahoo called Nova, which pushes continually-arriving data through graphs of Pig programs executing on Hadoop clusters. (Pig is a structured dataflow language and runtime for the Hadoop map-reduce system.)

We present a spreadsheet approach to simplifying the process of managing, programming, and interacting with sensor networks and visualizing, archiving and retrieving sensor data. An Excel spreadsheet prototype has been built to demonstrate the idea. This environment provides Excel users, who are already familiar with spreadsheet applications, a convenient and powerful tool for programming and data analysis. We discuss the architecture of this prototype and our experience in implementing the tool. We show two different classes of sensor-net applications built using this platform. We also present performance data on the scalability of the tool with respect to data rate and number of data streams.

No abstract available.

Due to the dynamic and ephemeral nature of ubiquitous computing (ubicomp) environments, it is necessary that application software in ubicomp environments is situation-aware (SA) and should be adaptable to both users' situation changes and the requirement changes. Reconfigurable context-sensitive middleware (RCSM) has been developed to provide development and runtime support for SA software in ubicomp environments, but do not provide runtime support for on-demand context acquisition, action scheduling and dynamic reconfiguration of SA software, and development support that maximize the reusability. In this work, the development and runtime support provided by RCSM have been substantially expanded to greatly simplify the development of situation-aware application software, and to achieve reusability and runtime reconfigurability simultaneously.