We propose a novel learning-based formulation for visual localization of vehicles that can operate in real-time in city-scale environments. Visual localization algorithms determine the position and orientation from which an image has been captured, using a set of geo-referenced images or a 3D scene representation. Our new localization paradigm, named Implicit Pose Encoding (ImPosing), embeds images and camera poses into a common latent representation with 2 separate neural networks, such that we can compute a similarity score for each image-pose pair. By evaluating candidates through the latent space in a hierarchical manner, the camera position and orientation are not directly regressed but incrementally refined. Very large environments force competitors to store gigabytes of map data, whereas our method is very compact independently of the reference database size. In this paper, we describe how to effectively optimize our learned modules, how to combine them to achieve real-time localization, and demonstrate results on diverse large scale scenarios that significantly outperform prior work in accuracy and computational efficiency.
The Emotion-Cause Pair Extraction (ECPE) task aims to extract emotions and causes as pairs from documents. We observe that the relative distance distribution of emotions and causes is extremely imbalanced in the typical ECPE dataset. Existing methods have set a fixed size window to capture relations between neighboring clauses. However, they neglect the effective semantic connections between distant clauses, leading to poor generalization ability towards position-insensitive data. To alleviate the problem, we propose a novel Multi-Granularity Semantic Aware Graph model (MGSAG) to incorporate fine-grained and coarse-grained semantic features jointly, without regard to distance limitation. In particular, we first explore semantic dependencies between clauses and keywords extracted from the document that convey fine-grained semantic features, obtaining keywords enhanced clause representations. Besides, a clause graph is also established to model coarse-grained semantic relations between clauses. Experimental results indicate that MGSAG surpasses the existing state-of-the-art ECPE models. Especially, MGSAG outperforms other models significantly in the condition of position-insensitive data.
Robot navigation in dynamic environments shared with humans is an important but challenging task, which suffers from performance deterioration as the crowd grows. In this paper, multi-subgoal robot navigation approach based on deep reinforcement learning is proposed, which can reason about more comprehensive relationships among all agents (robot and humans). Specifically, the next position point is planned for the robot by introducing history information and interactions in our work. Firstly, based on subgraph network, the history information of all agents is aggregated before encoding interactions through a graph neural network, so as to improve the ability of the robot to anticipate the future scenarios implicitly. Further consideration, in order to reduce the probability of unreliable next position points, the selection module is designed after policy network in the reinforcement learning framework. In addition, the next position point generated from the selection module satisfied the task requirements better than that obtained directly from the policy network. The experiments demonstrate that our approach outperforms state-of-the-art approaches in terms of both success rate and collision rate, especially in crowded human environments.
Real-time safety systems are crucial components of intelligent vehicles. This paper introduces a prediction-based collision risk assessment approach on highways. Given a point mass vehicle dynamics system, a stochastic forward reachable set considering two-dimensional motion with vehicle state probability distributions is firstly established. We then develop an acceleration prediction model, which provides multi-modal probabilistic acceleration distributions to propagate vehicle states. The collision probability is calculated by summing up the probabilities of the states where two vehicles spatially overlap. Simulation results show that the prediction model has superior performance in terms of vehicle motion position errors, and the proposed collision detection approach is agile and effective to identify the collision in cut-in crash events.
We propose a system that transcribes the conversation of a typical meeting scenario that is captured by a set of initially unsynchronized microphone arrays at unknown positions. It consists of subsystems for signal synchronization, including both sampling rate and sampling time offset estimation, diarization based on speaker and microphone array position estimation, multi-channel speech enhancement, and automatic speech recognition. With the estimated diarization information, a spatial mixture model is initialized that is used to estimate beamformer coefficients for source separation. Simulations show that the speech recognition accuracy can be improved by synchronizing and combining multiple distributed microphone arrays compared to a single compact microphone array. Furthermore, the proposed informed initialization of the spatial mixture model delivers a clear performance advantage over random initialization.
This study presents an autonomous orbit determination system based on crosslink radiometric measurements applied to a future lunar CubeSat mission to clearly highlight its advantages with respect to existing ground-based navigation strategies. This work is based on the Linked Autonomous Interplanetary Satellite Orbit Navigation (LiAISON) method which provides an autonomous navigation solution solely using satellite-to-satellite measurements, such as range and/or range-rate, to estimate absolute spacecraft states when at least one of the involved spacecraft has an orbit with a unique size, shape, and orientation. The lunar vicinity is a perfect candidate for this type of application due to the asymmetrical gravity field: the selected lunar mission, an Earth-Moon L2 (EML2) Halo orbiter, has an inter-satellite link between a lunar elliptical frozen orbiter. Simulation results show that, even in case of high-measurement errors (in the order of 100 m, 1 sigma, the navigation filter estimates the true states of spacecraft at EML2 with an error in the order of 500 m for position, and 2 mm/s for velocity, respectively and the elliptical lunar frozen orbiter states can be estimated in the order of 100 m for position and 1 cm/s for velocity, respectively. This study shows that range-only measurements provide better state estimation than range-rate-only measurements for this specific situation. Different bias handling strategies are also investigated. It has been found that even a less accurate ranging method, such as data-aided ranging, provides a sufficient orbit determination solution. This would simplify the communication system design for the selected CubeSat mission. The most observable states are found to be position states of the lunar orbiter via the observability analysis. In addition, the best tracking windows are also investigated for the selected mission scenario.
We consider 3-dimensional (3D) visible light positioning (VLP) based on smartphone camera in an indoor scenario. Based on the positioning model in the quantized pixel-domain, we characterize the 3D normalized positioning error metric (NPEM) through the partial derivative of the positioning function, and evaluate the NPEM for horizontal and non-horizontal receiver camera positions. Moreover, under horizontal receiver terminal position, we explore the relationship between the NPEM and the light-emitting diode (LED) cell layout, approximate the relationship between the NPEM and the number of LEDs captured by the camera, and evaluate the approximation accuracy according to the simulated positioning error. Based on the approximation results, we optimize the LED transmitter cell layout to minimize NPEM assuming structured square cell layouts with certain distance parameters.
The growing inequality in gig work between workers and platforms has become a critical social issue as gig work plays an increasingly prominent role in the future of work. The AI inequality is caused by (1) the technology divide in who has access to AI technologies in gig work; and (2) the data divide in who owns the data in gig work leads to unfair working conditions, growing pay gap, neglect of workers' diverse preferences, and workers' lack of trust in the platforms. In this position paper, we argue that a bottom-up approach that empowers individual workers to access AI-enabled work planning support and share data among a group of workers through a network of end-user-programmable intelligent assistants is a practical way to bridge AI inequality in gig work under the current paradigm of privately owned platforms. This position paper articulates a set of research challenges, potential approaches, and community engagement opportunities, seeking to start a dialogue on this important research topic in the interdisciplinary CHIWORK community.
Reconfigurable intelligent surfaces (RISs) have attracted enormous interest thanks to their ability to overcome line-of-sight blockages in mmWave systems, enabling in turn accurate localization with minimal infrastructure. Less investigated are however the benefits of exploiting RIS with suitably designed beamforming strategies for optimized localization and synchronization performance. In this paper, a novel low-complexity method for joint localization and synchronization based on an optimized design of the base station (BS) active precoding and RIS passive phase profiles is proposed, for the challenging case of a single-antenna receiver. The theoretical position error bound is first derived and used as metric to jointly optimize the BS-RIS beamforming, assuming a priori knowledge of the user position. By exploiting the low-dimensional structure of the solution, a novel codebook-based robust design strategy with optimized beam power allocation is then proposed, which provides low-complexity while taking into account the uncertainty on the user position. Finally, a reduced-complexity maximum-likelihood based estimation procedure is devised to jointly recover the user position and the synchronization offset. Extensive numerical analysis shows that the proposed joint BS-RIS beamforming scheme provides enhanced localization and synchronization performance compared to existing solutions, with the proposed estimator attaining the theoretical bounds even at low signal-to-noise-ratio and in the presence of additional uncontrollable multipath propagation.
Attribute-based Controlled Text Generation (CTG) refers to generating sentences that satisfy desirable attributes (e.g., emotions and topics). Existing works often utilize fine-tuning or resort to extra attribute classifiers, yet suffer from storage and inference time increases. To address these concerns, we explore attribute-based CTG in a prompt-based manner. In short, the proposed Tailor represents each attribute as a pre-trained continuous vector (i.e., single-attribute prompt) and guides the generation of a fixed PLM switch to a pre-specified attribute. We experimentally find that these prompts can be simply concatenated as a whole to multi-attribute CTG without any re-training, yet raises problems of fluency decrease and position sensitivity. To this end, Tailor provides a multi-attribute prompt mask and a re-indexing position-ids sequence to bridge the gap between the training (one prompt for each task) and testing stage (concatenating more than one prompt). To further enhance such single-attribute prompt combinations, Tailor also introduces a trainable prompt connector, which can be concatenated with any two single-attribute prompts to multi-attribute text generation. Experiments on 11 attribute-specific generation tasks demonstrate strong performances of Tailor on both single-attribute and multi-attribute CTG, with 0.08\% training parameters of a GPT-2.
A method of finding and classifying various components and objects in a design diagram, drawing, or planning layout is proposed. The method automatically finds the objects present in a legend table and finds their position, count and related information with the help of multiple deep neural networks. The method is pre-trained on several drawings or design templates to learn the feature set that may help in representing the new templates. For a template not seen before, it does not require any training with template dataset. The proposed method may be useful in multiple industry applications such as design validation, object count, connectivity of components, etc. The method is generic and domain independent.
Simile interpretation (SI) and simile generation (SG) are challenging tasks for NLP because models require adequate world knowledge to produce predictions. Previous works have employed many hand-crafted resources to bring knowledge-related into models, which is time-consuming and labor-intensive. In recent years, pre-trained language models (PLMs) based approaches have become the de-facto standard in NLP since they learn generic knowledge from a large corpus. The knowledge embedded in PLMs may be useful for SI and SG tasks. Nevertheless, there are few works to explore it. In this paper, we probe simile knowledge from PLMs to solve the SI and SG tasks in the unified framework of simile triple completion for the first time. The backbone of our framework is to construct masked sentences with manual patterns and then predict the candidate words in the masked position. In this framework, we adopt a secondary training process (Adjective-Noun mask Training) with the masked language model (MLM) loss to enhance the prediction diversity of candidate words in the masked position. Moreover, pattern ensemble (PE) and pattern search (PS) are applied to improve the quality of predicted words. Finally, automatic and human evaluations demonstrate the effectiveness of our framework in both SI and SG tasks.
In a two-period financial market where a stock is traded dynamically and European options at maturity are traded statically, we study the so-called martingale Schr\"odinger bridge Q*; that is, the minimal-entropy martingale measure among all models calibrated to option prices. This minimization is shown to be in duality with an exponential utility maximization over semistatic portfolios. Under a technical condition on the physical measure P, we show that an optimal portfolio exists and provides an explicit solution for Q*. This result overcomes the remarkable issue of non-closedness of semistatic strategies discovered by Acciaio, Larsson and Schachermayer. Specifically, we exhibit a dense subset of calibrated martingale measures with particular properties to show that the portfolio in question has a well-defined and integrable option position.
Chinese BERT models achieve remarkable progress in dealing with grammatical errors of word substitution. However, they fail to handle word insertion and deletion because BERT assumes the existence of a word at each position. To address this, we present a simple and effective Chinese pretrained model. The basic idea is to enable the model to determine whether a word exists at a particular position. We achieve this by introducing a special token \texttt{[null]}, the prediction of which stands for the non-existence of a word. In the training stage, we design pretraining tasks such that the model learns to predict \texttt{[null]} and real words jointly given the surrounding context. In the inference stage, the model readily detects whether a word should be inserted or deleted with the standard masked language modeling function. We further create an evaluation dataset to foster research on word insertion and deletion. It includes human-annotated corrections for 7,726 erroneous sentences. Results show that existing Chinese BERT performs poorly on detecting insertion and deletion errors. Our approach significantly improves the F1 scores from 24.1\% to 78.1\% for word insertion and from 26.5\% to 68.5\% for word deletion, respectively.
Vision-based bronchoscopy (VB) models require the registration of the virtual lung model with the frames from the video bronchoscopy to provide effective guidance during the biopsy. The registration can be achieved by either tracking the position and orientation of the bronchoscopy camera or by calibrating its deviation from the pose (position and orientation) simulated in the virtual lung model. Recent advances in neural networks and temporal image processing have provided new opportunities for guided bronchoscopy. However, such progress has been hindered by the lack of comparative experimental conditions. In the present paper, we share a novel synthetic dataset allowing for a fair comparison of methods. Moreover, this paper investigates several neural network architectures for the learning of temporal information at different levels of subject personalization. In order to improve orientation measurement, we also present a standardized comparison framework and a novel metric for camera orientation learning. Results on the dataset show that the proposed metric and architectures, as well as the standardized conditions, provide notable improvements to current state-of-the-art camera pose estimation in video bronchoscopy.
This position paper offers a framework to think about how to better involve human influence in algorithmic decision-making of contentious public policy issues. Drawing from insights in communication literature, we introduce a "public(s)-in-the-loop" approach and enumerates three features that are central to this approach: publics as plural political entities, collective decision-making through deliberation, and the construction of publics. It explores how these features might advance our understanding of stakeholder participation in AI design in contentious public policy domains such as recidivism prediction. Finally, it sketches out part of a research agenda for the HCI community to support this work.
To achieve accurate and robust object detection in the real-world scenario, various forms of images are incorporated, such as color, thermal, and depth. However, multimodal data often suffer from the position shift problem, i.e., the image pair is not strictly aligned, making one object has different positions in different modalities. For the deep learning method, this problem makes it difficult to fuse multimodal features and puzzles the convolutional neural network (CNN) training. In this article, we propose a general multimodal detector named aligned region CNN (AR-CNN) to tackle the position shift problem. First, a region feature (RF) alignment module with adjacent similarity constraint is designed to consistently predict the position shift between two modalities and adaptively align the cross-modal RFs. Second, we propose a novel region of interest (RoI) jitter strategy to improve the robustness to unexpected shift patterns. Third, we present a new multimodal feature fusion method that selects the more reliable feature and suppresses the less useful one via feature reweighting. In addition, by locating bounding boxes in both modalities and building their relationships, we provide novel multimodal labeling named KAIST-Paired. Extensive experiments on 2-D and 3-D object detection, RGB-T, and RGB-D datasets demonstrate the effectiveness and robustness of our method.
Recent advances in deep learning and computer vision offer an excellent opportunity to investigate high-level visual analysis tasks such as human localization and human pose estimation. Although the performance of human localization and human pose estimation has significantly improved in recent reports, they are not perfect and erroneous localization and pose estimation can be expected among video frames. Studies on the integration of these techniques into a generic pipeline that is robust to noise introduced from those errors are still lacking. This paper fills the missing study. We explored and developed two working pipelines that suited the visual-based positioning and pose estimation tasks. Analyses of the proposed pipelines were conducted on a badminton game. We showed that the concept of tracking by detection could work well, and errors in position and pose could be effectively handled by a linear interpolation technique using information from nearby frames. The results showed that the Visual-based Positioning and Pose Estimation could deliver position and pose estimations with good spatial and temporal resolutions.
Adaptive and intelligent user interfaces have been proposed as a critical component of a successful extended reality (XR) system. In particular, a predictive system can make inferences about a user and provide them with task-relevant recommendations or adaptations. However, we believe such adaptive interfaces should carefully consider the overall \emph{cost} of interactions to better address uncertainty of predictions. In this position paper, we discuss a computational approach to adapt XR interfaces, with the goal of improving user experience and performance. Our novel model, applied to menu selection tasks, simulates user interactions by considering both cognitive and motor costs. In contrast to greedy algorithms that adapt based on predictions alone, our model holistically accounts for costs and benefits of adaptations towards adapting the interface and providing optimal recommendations to the user.
Since 2017, the Transformer-based models play critical roles in various downstream Natural Language Processing tasks. However, a common limitation of the attention mechanism utilized in Transformer Encoder is that it cannot automatically capture the information of word order, so explicit position embeddings are generally required to be fed into the target model. In contrast, Transformer Decoder with the causal attention masks is naturally sensitive to the word order. In this work, we focus on improving the position encoding ability of BERT with the causal attention masks. Furthermore, we propose a new pre-trained language model DecBERT and evaluate it on the GLUE benchmark. Experimental results show that (1) the causal attention mask is effective for BERT on the language understanding tasks; (2) our DecBERT model without position embeddings achieve comparable performance on the GLUE benchmark; and (3) our modification accelerates the pre-training process and DecBERT w/ PE achieves better overall performance than the baseline systems when pre-training with the same amount of computational resources.