In this work, we consider the problem of estimating the 3D position of multiple humans in a scene as well as their body shape and articulation from a single RGB video recorded with a static camera. In contrast to expensive marker-based or multi-view systems, our lightweight setup is ideal for private users as it enables an affordable 3D motion capture that is easy to install and does not require expert knowledge. To deal with this challenging setting, we leverage recent advances in computer vision using large-scale pre-trained models for a variety of modalities, including 2D body joints, joint angles, normalized disparity maps, and human segmentation masks. Thus, we introduce the first non-linear optimization-based approach that jointly solves for the absolute 3D position of each human, their articulated pose, their individual shapes as well as the scale of the scene. In particular, we estimate the scene depth and person unique scale from normalized disparity predictions using the 2D body joints and joint angles. Given the per-frame scene depth, we reconstruct a point-cloud of the static scene in 3D space. Finally, given the per-frame 3D estimates of the humans and scene point-cloud, we perform a space-time coherent optimization over the video to ensure temporal, spatial and physical plausibility. We evaluate our method on established multi-person 3D human pose benchmarks where we consistently outperform previous methods and we qualitatively demonstrate that our method is robust to in-the-wild conditions including challenging scenes with people of different sizes.
Classical good continuation for image curves is based on $2D$ position and orientation. It is supported by the columnar organization of cortex, by psychophysical experiments, and by rich models of (differential) geometry. Here we extend good continuation to stereo. We introduce a neurogeometric model, in which the parametrizations involve both spatial and orientation disparities. Our model provides insight into the neurobiology, suggesting an implicit organization for neural interactions and a well-defined $3D$ association field. Our model sheds light on the computations underlying the correspondence problem, and illustrates how good continuation in the world generalizes good continuation in the plane.
We consider linear time-invariant dynamic systems in the single-input, single-output (SISO) framework. In particular, we consider stabilization of an inverted pendulum on a cart using a force on the cart. This system is easy to stabilize with pendulum angle feedback. However, with cart position feedback it cannot be stabilized with stable and proper compensators. Here we demonstrate that with an additional compensator in a parallel feedforward loop, stabilization is possible with such compensators. Sensitivity to noise seems to be about 3 times worse than for the situation with angle feedback. For completeness, discussion is presented of compensator parameter choices, robustness, fragility and comparison with another control approach.
We design and implement lab experiments to evaluate the normative appeal of behavior arising from models of ambiguity-averse preferences. We report two main empirical findings. First, we demonstrate that behavior reflects an incomplete understanding of the problem, providing evidence that subjects do not act on the basis of preferences alone. Second, additional clarification of the decision making environment pushes subjects' choices in the direction of ambiguity aversion models, regardless of whether or not the choices are also consistent with subjective expected utility, supporting the position that subjects find such behavior normatively appealing.
Recent research has highlighted the utility of Planar Parallax Geometry in monocular depth estimation. However, its potential has yet to be fully realized because networks rely heavily on appearance for depth prediction. Our in-depth analysis reveals that utilizing flow-pretrain can optimize the network's usage of consecutive frame modeling, leading to substantial performance enhancement. Additionally, we propose Planar Position Embedding (PPE) to handle dynamic objects that defy static scene assumptions and to tackle slope variations that are challenging to differentiate. Comprehensive experiments on autonomous driving datasets, namely KITTI and the Waymo Open Dataset (WOD), prove that our Planar Parallax Network (PPNet) significantly surpasses existing learning-based methods in performance.
Existing point cloud learning methods aggregate features from neighbouring points relying on constructing graph in the spatial domain, which results in feature update for each point based on spatially-fixed neighbours throughout layers. In this paper, we propose a dynamic feature aggregation (DFA) method that can transfer information by constructing local graphs in the feature domain without spatial constraints. By finding k-nearest neighbors in the feature domain, we perform relative position encoding and semantic feature encoding to explore latent position and feature similarity information, respectively, so that rich local features can be learned. At the same time, we also learn low-dimensional global features from the original point cloud for enhancing feature representation. Between DFA layers, we dynamically update the constructed local graph structure, so that we can learn richer information, which greatly improves adaptability and efficiency. We demonstrate the superiority of our method by conducting extensive experiments on point cloud classification and segmentation tasks. Implementation code is available: https://github.com/jiamang/DFA.
The forecast of electrical loads is essential for the planning and operation of the power system. Recently, advances in deep learning have enabled more accurate forecasts. However, deep neural networks are prone to adversarial attacks. Although most of the literature focuses on integrity-based attacks, this paper proposes availability-based adversarial attacks, which can be more easily implemented by attackers. For each forecast instance, the availability attack position is optimally solved by mixed-integer reformulation of the artificial neural network. To tackle this attack, an adversarial training algorithm is proposed. In simulation, a realistic load forecasting dataset is considered and the attack performance is compared to the integrity-based attack. Meanwhile, the adversarial training algorithm is shown to significantly improve robustness against availability attacks. All codes are available at https://github.com/xuwkk/AAA_Load_Forecast.
Graph matching can be formalized as a combinatorial optimization problem, where there are corresponding relationships between pairs of nodes that can be represented as edges. This problem becomes challenging when there are potential ambiguities present due to nodes and edges with high similarity, and there is a need to find accurate results for similar content matching. In this paper, we introduce a novel end-to-end neural network that can map the linear assignment problem into a high-dimensional space augmented with node-level relative position information, which is crucial for improving the method's performance for similar content matching. Our model constructs the anchor set for the relative position of nodes and then aggregates the feature information of the target node and each anchor node based on a measure of relative position. It then learns the node feature representation by integrating the topological structure and the relative position information, thus realizing the linear assignment between the two graphs. To verify the effectiveness and generalizability of our method, we conduct graph matching experiments, including cross-category matching, on different real-world datasets. Comparisons with different baselines demonstrate the superiority of our method. Our source code is available under https://github.com/anonymous.
We propose a novel approach for monocular 3D object detection by leveraging local perspective effects of each object. While the global perspective effect shown as size and position variations has been exploited for monocular 3D detection extensively, the local perspectives has long been overlooked. We design a local perspective module to regress a newly defined variable named keyedge-ratios as the parameterization of the local shape distortion to account for the local perspective, and derive the object depth and yaw angle from it. Theoretically, this module does not rely on the pixel-wise size or position in the image of the objects, therefore independent of the camera intrinsic parameters. By plugging this module in existing monocular 3D object detection frameworks, we incorporate the local perspective distortion with global perspective effect for monocular 3D reasoning, and we demonstrate the effectiveness and superior performance over strong baseline methods in multiple datasets.
We consider the Bachelier model with linear price impact. Exponential utility indifference prices are studied for vanilla European options in the case where the investor is required to liquidate her position. Our main result is establishing a non-trivial scaling limit for a vanishing price impact which is inversely proportional to the risk aversion. We compute the limit of the corresponding utility indifference prices and find explicitly a family of portfolios which are asymptotically optimal.
The paper proposes various strategies for sampling text data when performing automatic sentence classification for the purpose of detecting missing bibliographic links. We construct samples based on sentences as semantic units of the text and add their immediate context which consists of several neighboring sentences. We examine a number of sampling strategies that differ in context size and position. The experiment is carried out on the collection of STEM scientific papers. Including the context of sentences into samples improves the result of their classification. We automatically determine the optimal sampling strategy for a given text collection by implementing an ensemble voting when classifying the same data sampled in different ways. Sampling strategy taking into account the sentence context with hard voting procedure leads to the classification accuracy of 98% (F1-score). This method of detecting missing bibliographic links can be used in recommendation engines of applied intelligent information systems.
The Artemis program requires robotic and crewed lunar rovers for resource prospecting and exploitation, construction and maintenance of facilities, and human exploration. These rovers must support navigation for 10s of kilometers (km) from base camps. A lunar science rover mission concept - Endurance-A, has been recommended by the new Decadal Survey as the highest priority medium-class mission of the Lunar Discovery and Exploration Program, and would be required to traverse approximately 2000 km in the South Pole-Aitkin (SPA) Basin, with individual drives of several kilometers between stops for downlink. These rover mission scenarios require functionality that provides onboard, autonomous, global position knowledge ( aka absolute localization). However, planetary rovers have no onboard global localization capability to date; they have only used relative localization, by integrating combinations of wheel odometry, visual odometry, and inertial measurements during each drive to track position relative to the start of each drive. In this work, we summarize recent developments from the LunarNav project, where we have developed algorithms and software to enable lunar rovers to estimate their global position and heading on the Moon with a goal performance of position error less than 5 meters (m) and heading error less than 3-degree, 3-sigma, in sunlit areas. This will be achieved autonomously onboard by detecting craters in the vicinity of the rover and matching them to a database of known craters mapped from orbit. The overall technical framework consists of three main elements: 1) crater detection, 2) crater matching, and 3) state estimation. In previous work, we developed crater detection algorithms for three different sensing modalities. Our results suggest that rover localization with an error less than 5 m is highly probable during daytime operations.
First passage time (FPT) is the time a particle, subject to some stochastic process, hits or crosses a closed surface for the very first time. $\tau$-leaping methods are a class of stochastic algorithms in which, instead of simulating every single reaction, many reactions are ``leaped" over in order to shorten the computing time. In this paper we developed a $\tau$-leaping method for computing the FPT and position in arbitrary volumes for a Brownian particle governed by the Langevin equation. The $\tau$-leaping method proposed here works as follows. A sphere is inscribed within the volume of interest (VOI) centered at the initial particle's location. On this sphere, the FPT is sampled, as well as the position, which becomes the new initial position. Then, another sphere, centered at this new location, is inscribed. This process continues until the sphere becomes smaller than some minimal radius $R_{\text{min}}$. When this occurs, the $\tau$-leaping switches to the conventional Monte Carlo, which runs until the particle either crosses the surface of the VOI or finds its way to a position where a sphere of radius $>R_{\text{min}}$ can be inscribed. The switching between $\tau$-leaping and MC continues until the particle crosses the surface of the VOI. The size of this radius depends on the system parameters and on one's notion of accuracy: the larger this radius the more accurate the $\tau$-leaping method, but also less efficient. This trade off between accuracy and efficiency is discussed. For two VOI, the $\tau$-leaping method is shown to be accurate and more efficient than MC by at least a factor of 10 and up to a factor of about 110. However, while MC becomes exponentially slower with increasing VOI, the efficiency of the $\tau$-leaping method remains relatively unchanged. Thus, the $\tau$-leaping method can potentially be many orders of magnitude more efficient than MC.
Many multi-sensor navigation systems urgently demand accurate positioning initialization from global navigation satellite systems (GNSSs) in challenging static scenarios. However, ground blockages against line-of-sight (LOS) signal reception make it difficult for GNSS users. Steering local codes in GNSS basebands is a desiring way to correct instantaneous signal phase misalignment, efficiently gathering useful signal power and increasing positioning accuracy. Besides, inertial navigation systems (INSs) have been used as a well-complementary dead reckoning (DR) sensor for GNSS receivers in kinematic scenarios resisting various interferences since early. But little work focuses on the case of whether the INS can improve GNSS receivers in static scenarios. Thus, this paper proposes an enhanced navigation system deeply integrated with low-cost INS solutions and GNSS high-accuracy carrier-based positioning. First, an absolute code phase is predicted from base station information, and integrated solution of the INS DR and real-time kinematic (RTK) results through an extended Kalman filter (EKF). Then, a numerically controlled oscillator (NCO) leverages the predicted code phase to improve the alignment between instantaneous local code phases and received ones. The proposed algorithm is realized in a vector-tracking GNSS software-defined radio (SDR). Real-world experiments demonstrate the proposed SDR regarding estimating time-of-arrival (TOA) and positioning accuracy.
This position paper discusses the problem of multilingual evaluation. Using simple statistics, such as average language performance, might inject linguistic biases in favor of dominant language families into evaluation methodology. We argue that a qualitative analysis informed by comparative linguistics is needed for multilingual results to detect this kind of bias. We show in our case study that results in published works can indeed be linguistically biased and we demonstrate that visualization based on URIEL typological database can detect it.
Page layout analysis is a fundamental step in document processing which enables to segment a page into regions of interest. With highly complex layouts and mixed scripts, scholarly commentaries are text-heavy documents which remain challenging for state-of-the-art models. Their layout considerably varies across editions and their most important regions are mainly defined by semantic rather than graphical characteristics such as position or appearance. This setting calls for a comparison between textual, visual and hybrid approaches. We therefore assess the performances of two transformers (LayoutLMv3 and RoBERTa) and an objection-detection network (YOLOv5). If results show a clear advantage in favor of the latter, we also list several caveats to this finding. In addition to our experiments, we release a dataset of ca. 300 annotated pages sampled from 19th century commentaries.
In person re-identification (ReID) tasks, many works explore the learning of part features to improve the performance over global image features. Existing methods explicitly extract part features by either using a hand-designed image division or keypoints obtained with external visual systems. In this work, we propose to learn Discriminative implicit Parts (DiPs) which are decoupled from explicit body parts. Therefore, DiPs can learn to extract any discriminative features that can benefit in distinguishing identities, which is beyond predefined body parts (such as accessories). Moreover, we propose a novel implicit position to give a geometric interpretation for each DiP. The implicit position can also serve as a learning signal to encourage DiPs to be more position-equivariant with the identity in the image. Lastly, an additional DiP weighting is introduced to handle the invisible or occluded situation and further improve the feature representation of DiPs. Extensive experiments show that the proposed method achieves state-of-the-art performance on multiple person ReID benchmarks.
This paper proposes a novel sequence-to-sequence (seq2seq) model with a musical note position-aware attention mechanism for singing voice synthesis (SVS). A seq2seq modeling approach that can simultaneously perform acoustic and temporal modeling is attractive. However, due to the difficulty of the temporal modeling of singing voices, many recent SVS systems with an encoder-decoder-based model still rely on explicitly on duration information generated by additional modules. Although some studies perform simultaneous modeling using seq2seq models with an attention mechanism, they have insufficient robustness against temporal modeling. The proposed attention mechanism is designed to estimate the attention weights by considering the rhythm given by the musical score. Furthermore, several techniques are also introduced to improve the modeling performance of the singing voice. Experimental results indicated that the proposed model is effective in terms of both naturalness and robustness of timing.
While inferring common actor states (such as position or velocity) is an important and well-explored task of the perception system aboard a self-driving vehicle (SDV), it may not always provide sufficient information to the SDV. This is especially true in the case of active emergency vehicles (EVs), where light-based signals also need to be captured to provide a full context. We consider this problem and propose a sequential methodology for the detection of active EVs, using an off-the-shelf CNN model operating at a frame level and a downstream smoother that accounts for the temporal aspect of flashing EV lights. We also explore model improvements through data augmentation and training with additional hard samples.
Referring image segmentation aims to segment the target object described by a given natural language expression. Typically, referring expressions contain complex relationships between the target and its surrounding objects. The main challenge of this task is to understand the visual and linguistic content simultaneously and to find the referred object accurately among all instances in the image. Currently, the most effective way to solve the above problem is to obtain aligned multi-modal features by computing the correlation between visual and linguistic feature modalities under the supervision of the ground-truth mask. However, existing paradigms have difficulty in thoroughly understanding visual and linguistic content due to the inability to perceive information directly about surrounding objects that refer to the target. This prevents them from learning aligned multi-modal features, which leads to inaccurate segmentation. To address this issue, we present a position-aware contrastive alignment network (PCAN) to enhance the alignment of multi-modal features by guiding the interaction between vision and language through prior position information. Our PCAN consists of two modules: 1) Position Aware Module (PAM), which provides position information of all objects related to natural language descriptions, and 2) Contrastive Language Understanding Module (CLUM), which enhances multi-modal alignment by comparing the features of the referred object with those of related objects. Extensive experiments on three benchmarks demonstrate our PCAN performs favorably against the state-of-the-art methods. Our code will be made publicly available.