Search

Due to the visual properties of reflection and refraction, RGB-D cameras cannot accurately capture the depth of transparent objects, leading to incomplete depth maps. To fill in the missing points, recent studies tend to explore new visual features and design complex networks to reconstruct the depth, however, these approaches tremendously increase computation, and the correlation of different visual features remains a problem. To this end, we propose an efficient depth completion network named DistillGrasp which distillates knowledge from the teacher branch to the student branch. Specifically, in the teacher branch, we design a position correlation block (PCB) that leverages RGB images as the query and key to search for the corresponding values, guiding the model to establish correct correspondence between two features and transfer it to the transparent areas. For the student branch, we propose a consistent feature correlation module (CFCM) that retains the reliable regions of RGB images and depth maps respectively according to the consistency and adopts a CNN to capture the pairwise relationship for depth completion. To avoid the student branch only learning regional features from the teacher branch, we devise a distillation loss that not only considers the distance loss but also the object structure and edge information. Extensive experiments conducted on the ClearGrasp dataset manifest that our teacher network outperforms state-of-the-art methods in terms of accuracy and generalization, and the student network achieves competitive results with a higher speed of 48 FPS. In addition, the significant improvement in a real-world robotic grasping system illustrates the effectiveness and robustness of our proposed system., Comment: 10 pages, 5 figures

Large language models (LLMs) have been extensively studied for their abilities to generate convincing natural language sequences, however their utility for quantitative information retrieval is less well understood. In this paper we explore the feasibility of LLMs as a mechanism for quantitative knowledge retrieval to aid data analysis tasks such as elicitation of prior distributions for Bayesian models and imputation of missing data. We present a prompt engineering framework, treating an LLM as an interface to a latent space of scientific literature, comparing responses in different contexts and domains against more established approaches. Implications and challenges of using LLMs as 'experts' are discussed., Comment: 13 pages plus supplementary materials

An ever-increasing amount of social media content requires advanced AI-based computer programs capable of extracting useful information. Specifically, the extraction of health-related content from social media is useful for the development of diverse types of applications including disease spread, mortality rate prediction, and finding the impact of diverse types of drugs on diverse types of diseases. Language models are competent in extracting the syntactic and semantics of text. However, they face a hard time extracting similar patterns from social media texts. The primary reason for this shortfall lies in the non-standardized writing style commonly employed by social media users. Following the need for an optimal language model competent in extracting useful patterns from social media text, the key goal of this paper is to train language models in such a way that they learn to derive generalized patterns. The key goal is achieved through the incorporation of random weighted perturbation and contrastive learning strategies. On top of a unique training strategy, a meta predictor is proposed that reaps the benefits of 5 different language models for discriminating posts of social media text into non-health and health-related classes. Comprehensive experimentation across 3 public benchmark datasets reveals that the proposed training strategy improves the performance of the language models up to 3.87%, in terms of F1-score, as compared to their performance with traditional training. Furthermore, the proposed meta predictor outperforms existing health mention classification predictors across all 3 benchmark datasets.

This paper reports the findings of an experimental study on the problem of line-of-sight (LOS)/non-line-of-sight (NLOS) classification in an indoor environment. Specifically, we deploy a pair of NI 2901 USRP software-defined radios (SDR) in a large hall. The transmit SDR emits an unmodulated tone of frequency 10 KHz, on a center frequency of 2.4 GHz, using three different signal-to-noise ratios (SNR). The receive SDR constructs a dataset of pathloss measurements from the received signal as it moves across 15 equi-spaced positions on a 1D grid (for both LOS and NLOS scenarios). We utilize our custom dataset to estimate the pathloss parameters (i.e., pathloss exponent) using the least-squares method, and later, utilize the parameterized pathloss model to construct a binary hypothesis test for NLOS identification. Further, noting that the pathloss measurements slightly deviate from Gaussian distribution, we feed our custom dataset to four machine learning (ML) algorithms, i.e., linear support vector machine (SVM) and radial basis function SVM (RBF-SVM), linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and logistic regression (LR). It turns out that the performance of the ML algorithms is only slightly superior to the Neyman-Pearson-based binary hypothesis test (BHT). That is, the RBF-SVM classifier (the best performing ML classifier) and the BHT achieve a maximum accuracy of 88.24% and 87.46% for low SNR, 83.91% and 81.21% for medium SNR, and 87.38% and 86.65% for high SNR., Comment: 5 pages, 4 figures, submitted to an IEEE conference