Abstract

Brain stroke has become a significant burden on global health and thus we need remedies and prevention strategies to overcome this challenge. For this, the immediate identification of stroke and risk stratification is the primary task for clinicians. To aid expert clinicians, automated segmentation models are crucial. In this work, we consider the publicly available dataset ATLAS v2.0 to benchmark various end-to-end supervised U-Net style models. Specifically, we have benchmarked models on both 2D and 3D brain images and evaluated them using standard metrics. We have achieved the highest Dice score of 0.583 on the 2D transformer-based model and 0.504 on the 3D residual U-Net respectively. We have conducted the Wilcoxon test for 3D models to correlate the relationship between predicted and actual stroke volume.

Abstract

We study communication models for channels with erasures in which the erasure pattern can be controlled by an adversary with partial knowledge of the transmitted codeword. In particular, we design block codes for channels with binary inputs with an adversary who can erase a fraction p of the transmitted bits. We consider causal adversaries, who must choose to erase an input bit using knowledge of that bit and previously transmitted bits, and myopic adversaries, who can choose an erasure pattern based on observing the transmitted codeword through a binary erasure channel with random erasures. For both settings we design efficient (polynomial time) encoding and decoding algorithms that use randomization at the encoder only. Our constructions achieve capacity for the causal and "sufficiently myopic" models. For the "insufficiently myopic" adversary, the capacity is unknown, but existing converses show the capacity is zero for a range of parameters. For all parameters outside of that range, our construction achieves positive rates.

Abstract

AI is about learning algorithms and huge amounts of data and are drivers of economic growth -- what does this mean for the field of development studies? Can we re-orient to twin AI studies and development theory and practice to generate how development challenges are identified and researched? To do this a good grasp is needed of AI internal mechanisms and outcomes in addressing development issues -- this argument will be developed through a case study of the ADAS [Advanced Driver Assistance System] deployment in India. Over and above discussing the ADAS we bring an anthropological lens to understand the social context that surrounds the system. Focusing on bus drivers, we offer findings from a qualitative and ethnographic study of drivers in a collaborative effort to achieve road safety by deploying AI-driven technology and empowering stakeholders in the transport industry in India especially, bus drivers as critical actors in the city transport network.

Abstract

One crucial aspect of content organization is the creation of article outlines, which summarize the primary topics and subtopics covered in an article in a structured manner. This paper introduces a solution called XOutlineGen, which generates cross-lingual outlines for encyclopedic texts from reference articles. XOutlineGen uses the XWikiRef dataset, which consists of encyclopedic texts generated from reference articles and section titles. The dataset is enhanced with two new languages and three new domains, resulting in ∼92K articles. Our pipeline employs this dataset to train a two-step generation model, which takes the article title and set of references as inputs and produces the article outline.

Abstract

Researchers quantify vegetation characteristics using phytosociology studies. The data collected from vegetation surveys, such as tree girth and height measurements, are used to calculate various phytosociological attributes like frequency, density, basal cover, important value index, etc. In general, they are calculated using an Excel spreadsheet and a lengthy mathematical process. There are a few open-source or free software available to compute some of these parameters. However, they are not user friendly for most researchers. Also, these software primarily doesn’t compute basic phytosociological attributes. In view of these constraints and to ease researchers work, we have developed a simple tool, “PAT” Phytosociology Analysis Tool,” that computes the maximum attributes from the field collected vegetation data. The PAT GUI is simple. PAT computes results using standard formulas after users upload a csv file in the required format. Users can download the table's results and figures in PNG format.

Abstract

The collapse of buildings due to an earthquake is the leading cause of higher life loss and economic loss. it is not possible to prevent earthquake from occurring,but it disastrous effects can be minimized considerably through measure of scientific methods and understanding

Abstract

Multiple business scenarios require an automated generation of descriptive human-readable long text from structured input data, where the source is typically a high-resource language and the target is a low or medium resource language. We define the Cross-Lingual Fact to Long Text Generation (XFLT) as a novel natural language generation (NLG) task that involves generating descriptive and human-readable long text in a target language from structured input data (such as fact triples) in a source language. XFLT is challenging because of (a) hallucinatory nature of the state-of-the-art NLG models, (b) lack of good quality training data, and (c) lack of a suitable cross-lingual NLG metric. Unfortunately previous work focuses on different related problem settings (cross-lingual facts to short text or monolingual graph to text) and has made no efforts to handle hallucinations. In this paper, we contribute a novel dataset, XLALIGN with over 64,000 paragraphs across 12 different languages, and English facts. We propose a novel solution to the XFLT task which addresses these challenges by training multilingual Transformer-based encoder-decoder models with coverage prompts and grounded decoding. Further, it improves on the XFLT quality by defining task-specific reward functions and training on them using reinforcement learning. On XLALIGN, we compare this novel solution with several strong baselines using a new metric, cross-lingual PARENT. We also make our code and data publicly available

Abstract

Regular plantar pressure monitoring is required for diabetic, wheelchair-bound and bed-ridden individuals. It has been shown that peak plantar pressure for diabetics increases significantly and progressively, and requires regular monitoring for early detection and intervention. Sensor matrices of capacitative, resistive, and piezoelectric transducers are most commonly used for measuring areal pressure. Calibration for such medical devices is crucial for reliability and repeatable results. In this work, we elaborate on the calibration challenges of a 32×32 velostat-based sensor matrix, and demonstrate calibration strategies suitable for a manufactured product. We show that 2N parameters can be used to calibrate an N×N sensor matrix instead of 2N2 . We observe the accuracy of calibration of a 32×32 sensor matrix by comparing the total detected weight of 35 participants at a static load with their reported weight, with 4.2% mean error and 3.1% median error. For participants in the 50–85 kg weight range, the difference was always within ±5%. Further, upon drawing contour plots and using the pressure distribution information, we were able to distinguish participants with possible foot ulcers and flat feet.

Abstract

Pressure sensors are subjected to continuous force and stress that may affect the operation of the sensor in the long run. Reliability is a crucial factor that must be considered when designing and fabricating any sensor. It is essential to test the material used in the sensor to assess the reliability of the complete product. In this work, we report the long-term reliability of a flexible pressure sensor mat using a carbon-impregnated polymer, velostat, which is a flexible, light, and thin polymer composite material with piezoresistive properties. We focus on the analysis of the performance of a flexible pressure sensor array under long-term and repeated loading. Tests were performed every fortnight for 210 days. We have observed that the material characteristics of the velostat material change on repeated application of pressure up to a certain time frame. For a given loading, once the material settles, the change in resistance of the material becomes consistent for a given application of pressure. We have also analyzed the changes in the parameters associated with the 2-parameter model, and have analyzed the effect of crosstalk on the sensor matrix for different pitch lengths to select the best pitch that will give us the minimum crosstalk. We have observed that the error rate of the sensor pixels decreased by 53 percentage points in 210 days. The results obtained from the experimental tests for reliability reveal a practical possibility of implementing velostat-based pressure sensors in wearable and healthcare devices and provide steps to take while calibrating an as-fabricated velostat-based sensor.

Abstract

Abstract—Building 3D maps of the environment is central to robot navigation, planning, and interaction with objects in a scene. Most existing approaches that integrate semantic concepts with 3D maps largely remain confined to the closed-set setting: they can only reason about a finite set of concepts, pre-defined at training time. Further, these maps can only be queried using class labels, or in more recent work, using text prompts. We address both these issues with ConceptFusion, a scene representation that is: (i) fundamentally open-set, enabling reasoning beyond a closed set of concepts (ii) inherently multimodal, enabling a diverse range of possible queries to the 3D map, from language, to images, to audio, to 3D geometry, all working in concert. ConceptFusion leverages the open-set capabilities of today’s foundation models that have been pretrained on internet-scale data to reason about concepts across modalities such as natural language, images, and audio. We demonstrate that pixel-aligned open-set features can be fused into 3D maps via traditional SLAM and multi-view fusion approaches. This enables effective zero-shot spatial reasoning, not needing any additional training or finetuning, and retains long-tailed concepts better than supervised approaches, outperforming them by more than 40% margin on 3D IoU. We extensively evaluate ConceptFusion on a number of real-world datasets, simulated home environments, a real-world tabletop manipulation task, and an autonomous driving platform. We showcase new avenues for blending foundation models with 3D open-set multimodal mapping. We encourage the reader to view the demos on our project page: https://concept-fusion.github.io