Abstract

Accurately recognizing a revisited place is crucial for embodied agents to localize and navigate. This requires visual representations to be distinct, despite strong variations in camera viewpoint and scene appearance. Existing visual place recognition pipelines encode the whole image and search for matches. This poses a fundamental challenge in matching two images of the same place captured from different camera viewpoints: the similarity of what overlaps can be dominated by the dissimilarity of what does not overlap. We address this by encoding and searching for image segments instead of the whole images. We propose to use open-set image segmentation to decompose an image into `meaningful' entities (i.e., things and stuff). This enables us to create a novel image representation as a collection of multiple overlapping subgraphs connecting a segment with its neighboring segments, dubbed SuperSegment. Furthermore, to efficiently encode these SuperSegments into compact vector representations, we propose a novel factorized representation of feature aggregation. We show that retrieving these partial representations leads to significantly higher recognition recall than the typical whole image based retrieval. Our segments-based approach, dubbed SegVLAD, sets a new state-of-the-art in place recognition on a diverse selection of benchmark datasets, while being applicable to both generic and task-specialized image encoders. Finally, we demonstrate the potential of our method to revisit anything by evaluating our method on an object instance retrieval task, which bridges the two disparate areas of research: visual place recognition and object-goal navigation, through their common aim of recognizing goal objects specific to a place.

Abstract

Current algorithms for Maximal Frequent Subgraph (MFS) mining do not scale to large databases with more than 300k graphs. Frequency computation is commonly done using numerous subgraph isomorphism operations, which are computationally expensive. This paper explores a partition-based secondary memory algorithm, PMCS, that can make MFS mining for large databases viable. PMCS intelligently avoids redundant computations to perform frequency computation operations optimally. Its intermediate results are also small enough to fit in the main memory. Our results demonstrate that PMCS scales to databases of up to 1000k graphs with an average of 25-30 edges per graph.

Abstract

The Indian industrial sector has recorded a double-digit GDP growth of 10.3% in the first quarter of 2005-06. Substantial investments were made in the industrial sector, in general, and in heavy industries, in particular. Most of the industrial structures were constructed before 2000, and the buildings are in continuous production of goods. Due to the harsh chemical environment, the structural elements are getting distressed prior to the life span of the building. To avoid or reduce the shutdown period of the building, only damaged elements are preferred for strengthening. In the current study, a G+4 industrial building is considered. One side of the perimeter columns is damaged due to chemical spillage, which led to spalling and corrosion of columns. In the present study, the safety and efficacy of the building are reviewed using nonlinear static analysis in ETABS. The damaged columns are strengthened using concrete jacketing according to the Indian code. In case 1, only damaged columns at ground floor are strengthened as per the code provisions, and then the building is opened for use. In case 2, the building is strengthened at damaged columns for all floors and is analysed with nonlinear static analysis to compute the updated capacity of the column. Further, damage distribution is observed after the redistribution of moments due to stiffness modification in the building system. The decision to further strengthen is taken by observing the damage distribution and comparing updated capacity with the base shear of the region. The study demonstrated that the building with strengthening as per the base shear of the building with global retrofitting performed better than the building strengthened at ground floor columns.

Abstract

No Results Found

Abstract

Efficiently generating grasp poses tailored to specific regions of an object is vital for various robotic manipulation tasks, especially in a dual-arm setup. This scenario presents a significant challenge due to the complex geometries involved, requiring a deep understanding of the local geometry to generate grasps efficiently on the specified constrained regions. Existing methods only explore settings involving table-top/small objects and require augmented datasets to train, limiting their performance on complex objects. We propose CGDF: Constrained Grasp Diffusion Fields, a diffusion-based grasp generative model that generalizes to objects with arbitrary geometries, as well as generates dense grasps on the target regions. CGDF uses a part-guided diffusion approach that enables it to get high sample efficiency in constrained grasping without explicitly training on massive constraint-augmented datasets. We provide qualitative and quantitative comparisons using analytical metrics and in simulation, in both unconstrained and constrained settings to show that our method can generalize to generate stable grasps on complex objects, especially useful for dual-arm manipulation settings, while existing methods struggle to do so. More results, code and an extended version of the paper can be found on the project page: https://constrained-grasp-diffusion.github.io/

Abstract

This work presents a novel approach to identifying possible early accident-prone zones in a large city-scale road network using geo-tagged collision alert data from a vehicle fleet. The alert data has been collected for a year from 200 city buses installed with the Advanced Driver Assistance System (ADAS). To the best of our knowledge, no research paper has used ADAS alerts to identify the early accident prone zones. A nonparametric technique called Kernel Density Estimation (KDE) is employed to model the distribution of alert data across stratified time intervals. A novel recall-based measure is introduced to assess the degree of support provided by our density-based approach for existing, manually determined accident-prone zones (‘blackspots’) provided by civic authorities. This shows that our KDE approach significantly outperforms existing approaches in terms of the recall-based measure. Introducing a novel linear assignment Earth Mover Distance based measure to predict previously unidentified accident-prone zones. The results and findings support the feasibility of utilizing alert data from vehicle fleets to aid civic planners in assessing accident-zone trends and deploying traffic calming measures, thereby improving overall road safety and saving lives.

Abstract

Exploring and quantifying semantic relatedness is central to representing language and holds significant implications across various NLP tasks. While earlier NLP research primarily focused on semantic similarity, often within the English language context, we instead investigate the broader phenomenon of semantic relatedness. In this paper, we present SemRel, a new semantic relatedness dataset collection annotated by native speakers across 13 languages: Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Spanish, and Telugu. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia – regions characterised by a relatively limited availability of NLP resources. Each instance in the SemRel datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. The scores are obtained using a comparative annotation framework. We describe the data collection and annotation processes, challenges when building the datasets, baseline experiments, and their impact and utility in NLP.

Abstract

We present the first shared task on Semantic Textual Relatedness (STR). While earlier shared tasks primarily focused on semantic similarity, we instead investigate the broader phenomenon of semantic relatedness across 14 languages: Afrikaans, Algerian Arabic, Amharic, English, Hausa, Hindi, Indonesian, Kinyarwanda, Marathi, Moroccan Arabic, Modern Standard Arabic, Punjabi, Spanish, and Telugu. These languages originate from five distinct language families and are predominantly spoken in Africa and Asia – regions characterised by the relatively limited availability of NLP resources. Each instance in the datasets is a sentence pair associated with a score that represents the degree of semantic textual relatedness between the two sentences. Participating systems were asked to rank sentence pairs by their closeness in meaning (i.e., their degree of semantic relatedness) in the 14 languages in three main tracks: (a) supervised, (b) unsupervised, and (c) crosslingual. The task attracted 163 participants. We received 70 submissions in total (across all tasks) from 51 different teams, and 38 system description papers. We report on the best-performing systems as well as the most common and the most effective approaches for the three different tracks.

Abstract

The spot size variation due to mechanical jitter is calculated for the Hermite Gaussian (HG) incoherent beam combination system. It is observed that the spot size increases with an increase in mechanical jitter and higher modes HG beams experience less mechanical jitter effect compared to lower mode HG beams.

Abstract

Street lights are ubiquitous public infrastructure in urban areas that can be leveraged for smart city applications. In this paper, we present an electric vehicle charging station (EVCS) that communicates using Wi-SUN network integrated on a network of streetlights, and using oneM2M middleware. The proposed architecture follows the level 2 charging standards and the open charge point protocol (OCPP), ensuring compatibility and interoperability across diverse charging infrastructures. By seamlessly incorporating Wi-SUN network and oneM2M middleware, our architecture enables efficient communication and interaction with a wide array of devices and services. The utilization of the oneM2M platform further enhances the integration by establishing a seamless connection between the EVCS and the broader smart city infrastructure. We have fabricated a proof-of-concept system that consists of the EVCS connected to a Wi-SUN network integrated with the streetlight network in the institute campus. The performance of our solution was evaluated by measuring the latency associated with authentication and billing for EV users. We report the average latency observed over several iterations of charging to be 0.7±0.2 s (excluding the time required for charging). We also measured the Wi-SUN communication range in the campus environment with trees and buildings and found the maximum range to be around 370 m.