Abstract

Preserving cultural heritage is of paramount importance. In the domain of art restoration, developing a computer vision model capable of effectively restoring deteriorated images of art pieces was difficult but now we have good computer vision state-of-art. Traditional restoration methods are often time-consuming and require extensive expertise. The aim of this work is to design an automated solution based on computer vision models that can enhance and reconstruct degraded artworks, improving their visual quality while preserving their original characteristics and artifacts. The model should handle a diverse range of deterioration types, including but not limited to noise, blur, scratches, fading, and other common forms of degradation. We adapt the current state-of-art for the image super resolution based on Diffusion Model (DM) and fine-tune it for Image art restoration. Our result show that the instead of fine-tunning multiple different model for different kind of degradation, fine- tunning one super-resolution. We train it on the multiple datasets to make it robust.

Abstract

Speech systems are sensitive to accent variations. This is a challenge in India, which has numerous languages but few linguistic studies on pronunciation variation. The growing number of L2 English speakers reinforces the need to study accents and L1-L2 interactions. We investigate Indian English (IE) accents and report our observations on regional and shared features. Specifically, we observe phonemic variations and phonotactics in speakers’ native languages and apply this to their English pronunciations. We demonstrate the influence of 18 Indian languages on IE by comparing native language features with IE pronunciations obtained from literature studies and phonetically annotated speech. Hence, we validate Indian language influences on IE by justifying pronunciation rules from the perspective of Indian language phonology. We obtain a comprehensive description of generalised and region-specific IE characteristics, which facilitates accent adaptation of existing speech systems.

Abstract

Cocaine use disorder (CUD) is described as a compulsive urge to seek and consume cocaine despite the inimical consequences. MRI studies from different modalities have shown that CUD patients exhibit structural and/or functional connectivity pathology among several brain regions. Nevertheless, both connectivities are commonly studied and analyzed separately, which may potentially obscure its relationship between them, and with the clinical pathology. Here, we compare and contrast structural and functional brain networks in CUD patients and healthy controls (HC) using multimodal fusion. The sample consisted of 63 (8 females) CUD patients and 42 (9 females) healthy controls (HC), recruited as part of the SUDMEX CONN database. For this, we computed a battery of graph-based measures from multi-shell diffusion-weighted imaging and resting state fc-fMRI to quantify local and global connectivity. Then we used multimodal canonical component analysis plus joint independent component analysis (mCCA+jICA) to compare between techniques, and evaluate group differences and its association with clinical alteration. Unimodal results showed a striatal decrease in the participation coefficient, but applied supervised data fusion revealed other regions with cocaine-related alterations in joint functional communication. When performing multimodal fusion analysis, we observed a higher centrality of the interrelationship and a lower participation coefficient in patients with CUD. In contrast to the unimodal approach, the multimodal fusion method was able to reveal latent information about brain regions involved in impairment due to cocaine abuse. The present results could help in understanding the pathology of CUD in order to develop better pre-treatment/post-treatment intervention designs

Abstract

Events and states have gained importance in NLP and information retrieval for being semantically rich temporal and spatial information indicators. Event causality helps us identify which events are necessary for another event to occur. The cause-effect event pairs can be relevant for multiple NLP tasks like question answering, summarization, etc. Multiple efforts have been made to identify causal events in documents but very little work has been done in this field in the Hindi language. We create an annotated corpus for detecting and classifying causal event relations on top of the Hindi Timebank (Goel et al., 2020), the ‘Hindi Causal Timebank’ (Hindi CTB). We introduce semantic causal relations like Purpose, Reason, and Enablement inspired from Bejan and Harabagiu (2008)’s annotation scheme and add some special cases particular to Hindi language.

Abstract

Social media was one of the most popular forms of communication among young people of a certain class demographic during the pandemic. Consequently, crucial debates and discussions about the pandemic crisis itself have also developed on social media platforms, making them a great primary source to study the experiences of the LGBTQ+ community during the pandemic. We conducted research using LDA topic modelling on a sub-reddit from the Reddit platform to understand the nature of this discourse. LDA Topic modelling is useful in identifying patterns and themes in large volumes of unstructured data and is an appropriate tool for analyzing social media posts. The results from our preliminary research in this area suggest that institutions such as the health care system, justice system, and legislative system perpetuated systematic inequalities against LGBTQ+ communities during the pandemic, thereby adding to the pre-existing stigma against them during a global crisis. Data from subreddits such as ‘lgbt’ have also suggested a shift in tone in these discussions from the period during the pandemic to the period after the pandemic. We intend to now expand our analysis to other platforms, such as Twitter and Facebook to verify and qualify our approach and understanding of this problem. Researchers, such as Ahmed & Sifat (2021), Pandya & Redcay (2021) and Bhalla & Agarwal (2021), working on qualitative studies have given us enough evidence that proves that inequalities and violence against these communities had intensified during the pandemic. Although research has been done showing that COVID-19 pandemic correlates to an increased amount of discrimination towards the LGBTQ+ community, no research prior to this study has examined content available on Twitter, Facebook and Reddit using LDA topic modelling to understand this correlation. By conducting this research, we will also be able to offer an informed analysis of the effectiveness of computational tools in the study of gender and look at this population from a new perspective.

Abstract

In the twenty-first century, water quality monitoring is a growing challenge in inland water bodies. Therefore, monitoring the current state of the water bodies is highly important. The remote sensing data is important for effective monitoring that covers a large area. In recent decades, chlorophyll-a (as a proxy) has been a significant indicator of nutrient contamination and also indicates the qualitative status of the water bodies. Besides, the waterbody's temperature also plays an important role in the biological and chemical processes. Hence, the present study aims to detect the chlorophyll-a spread area and temperature of the Bhadra Reservoir, Tungabhadra River system, India. To explore the quality of the reservoir, Sentinel 2A and Landsat 8 satellite images were used for the years 2017 and 2018. The study results noticed that chlorophyll-a spread percent was higher in summer months compared with the water spread area of the reservoir, i.e., 74.5% in 2017 and 61.4% in 2018 from the total water spread area. The study also revealed that as the temperature rises, the chlorophyll-a spread also increases in the reservoir. It was also observed that the October month has a low chlorophyll-a spread, i.e., 32.04% in 2017 and 24.35% in 2018, because the one side reservoir is covered by forest area. Thus, there is a need to explore further to understand the impact of forest area on the reservoir and the temporal changes of the water quality parameters that decide the water bodies quality.

Abstract

The scalability of remote Internet of Things (IoT) deployments into verticals of smart cities relies primarily on the communication protocols and data platforms employed. Accordingly, the adoption of LoRaWAN, Sigfox, NB-IoT, etc. has risen in recent years. These protocols have been created to cater to the low power and long range requirements of IoT. In this paper, we evaluate one such protocol, Wi-SUN (Wireless Smart Ubiquitous Network), that additionally offers excellent redundancy owing to its use of a mesh topology. Wi-SUN is based on the IEEE 802.15.4g standard and uses license-free sub-GHz bands. However, IoT networks using any communication protocol require a knowledge of signal propagation and attenuation in real-world scenarios. Thus, we evaluated the path loss of WiSUN transmissions by analysing the RSSI values inside a typical city building and an urban outdoor habitat

Abstract

Warm and dry climate extremes have large impact on water resources, agriculture, ecosystems, and human health. The present study used high-resolution rainfall and temperature data at 0.25-degree resolution from 1951 to 2014 to investigate the occurrence of compound warm and dry extreme events across India. The simultaneous occurrence of temperature exceeding the 75th percentile and a Standardized Precipitation Evapotranspiration Index (SPEI) of less than –1 is considered to represent the compound extreme event of warm and dry. The study reveals a strong negative correlation between temperature and SPEI across most of the country except south-west of India. Analysis reveals that about 3.1% of increase in the spatial areal extent of Warm/Dry compound extreme events allover India. The insights derived from this research could significantly contribute to shaping climate change mitigation and adaptation strategies related to compound extreme events.

Abstract

Embodiments herein provide a system and method for automatically generating a sign language video from an input speech using the machine learning model. The method includes (i) extracting a plurality of spectrograms of an input speech by (a) encoding, using an encoder, a time domain series of the input speech to a frequency domain series, and (b) decoding, using a decoder, a plurality of tokens for time steps of the frequency domain series, (ii) generating a plurality of pose sequences for a current time step of the plurality of spectrograms using a first machine learning model, and (iii) automatically generating, using a discriminator of a second machine learning model, a sign language video for the input speech using the plurality of pose sequences and the plurality of spectrograms when the plurality of pose sequences are matched with corresponding the plurality of spectrograms that are extracted.

Abstract

The advances in Artificial Intelligence are creating new opportunities to improve people’s lives around the world, from business to healthcare, from lifestyle to education. For example, some systems profile the users using their demographic and behavioral characteristics to make certain domain-specific predictions. Often, such predictions impact the user’s life directly or indirectly (e.g., loan disbursement, determining insurance coverage, shortlisting applications, etc.). As a result, the concerns over such AI-enabled systems are also increasing. To address these concerns, such systems are mandated to be responsible: transparent, fair, and explainable to developers and end-users. In this paper, we present ComplAI, a unique framework to enable, observe, analyze and quantify explainability, robustness, performance, fairness, and model’s behavior in drift scenarios, and to provide a single Trust Factor that evaluates different supervised Machine Learning models not just from their ability to make correct predictions but from overall responsibility perspective. The framework helps users to (a) connect their models and enable explanations, (b) assess and visualize different aspects of the model such as robustness, drift susceptibility, and fairness, and (c) compare different models (from different model families or obtained through different hyperparameter settings) from an overall perspective thereby facilitating actionable recourse for improvement of the models. ComplAI is model agnostic and works with different supervised machine learning scenarios (i.e., Binary Classification, Multi-class Classification, and Regression) and frameworks (viz. scikit-learn, TensorFlow, etc.). It can be seamlessly integrated with any ML life-cycle framework. Thus, this already deployed framework aims to unify critical aspects of Responsible AI systems for regulating the development process of such real systems. The poster paper of ComplAI is also available