Abstract

n this paper, we try to tackle the problem of erroneous English-Hindi machine translation (MT) outputs due to the presence of the Verb Phrase Ellipsis (VPE) in English. The phenomenon of VPE is prominent in spoken English, and the antecedent to the ellipsis can come from previous sentences in a conversation as well. MT systems translate sentences as a whole and ignore the contextual information from the previous sentences. For these two reasons, spoken English-Hindi transla- tions suffer. We approached this problem by manually annotating 1200 two-person conversations that contain VPE and by studying how their resolution affects the translation qualities. Using our studies, we designed a rule-based system for the detection and resolution of VPE in English with the goal of improving their subsequent Hindi translation qualities. Our rule-based system is capable of the following: 1) Detection of VPE, 2) Resolution of Elided Head verb, 3) Resolution of Elided Head verb’s children, 4) Resolution of non-verbal predicates of a copula or a ’be’ main verb, 5) Modifying original sentence in the conversation with the resolved verb phrase. We also tested the system’s performance on VPE datasets outside of our annotated data. In this paper, we present our annotated corpus on conversational English VPE, our rule-based system to tackle VPE in the context of improving English-Hindi MT, the observations made as we designed this rule-based system and the performance-related observations of our system

Abstract

Across the globe, stone has been extensively used as a building construction material due to its local availability and high durability. The scenario for the construction of buildings in India is no more different than compared to other parts of the world. According to the 2011 Census of India, 43.48 million houses (~ 14% of the total number of houses) have stone as the predominant wall material. Owing to the fact of non-engineered construction, wide variations in terms of stone masonry construction can be seen in the Indian subcontinent, particularly in the Himalayan belt. In this study, a field survey is conducted to identify the seismic resistant features in traditional building practices which make use of stone for construction, e.g. Thathara construction, Dry stack construction, Koti Banal construction, etc. The percentage of these types of structures is decreasing due to an increase in urbanization, increasing housing demands, non-availability of traditional construction materials and skilled artesian for building new structures as well as costly repair and retrofitting techniques for damaged and existing structures. Thus, there is a dire need to safeguard these structures as if this trend continues, these buildings will become things of the past. In this paper, a review of structural configuration, roofing systems, foundation type and potential failure modes of these construction typologies are studied, and the effect of different retrofitting techniques that exist in the construction type on their expected seismic performance is discussed.

Abstract

Recurrent earthquakes in the past two decades have taken thousands of lives (e.g., the Kashmir earthquake killed some 80,000 people) and destroyed millions of homes and other buildings. This colossal loss of lives was not because of the earthquake but because people were inside buildings (e.g., stone masonry structures) that were highly vulnerable. The death toll would have been significantly lower, if the structures had been retrofitted with a small fraction of the cost of reconstruction. The past earthquakes have brought out many weaknesses in masonry design and construction practices as these structures behaved unsatisfactorily and sustained major damage. Hence, the safety and retrofitting of existing stone masonry houses are very important. Considerable research work has been directed toward evolving suitable methods of earthquake resistance in stone masonry houses.Despite the availability of such methods, masonry buildings have been damaged in the event of earthquakes, because of the following reasons: (i) lack of awareness, formal training, and technical knowledge in earthquake-resistant construction; (ii) lack of concern about seismic safety because of the infrequent occurrence of earthquakes; (iii) people lack financial resources to meet the earthquake-resistant requirement; and (iv) despite the availability of provisions and recommendations of earthquake-resistant measures to be applied on stone masonry buildings in the form of various codes, and these are rarely being implemented in actual practice. Therefore, in the present study, an overview is presented about how the retrofitting of existing stone masonry houses using different techniques can be implemented to reduce their seismic vulnerability and loss of life.

Abstract

Stone masonry walls have inherent weaknesses against the lateral forces of an earthquake. These weaknesses result in inadequate performance during earthquakes. Mistakes are commonly committed in the construction of stone walls, especially in the random rubble-type masonry. These mistakes further erode strength. When shaken, poorly constructed walls having inadequate interlocking between the inside and the outside faces (wythes), the faces begin to separate, resulting in rapid weakening of the wall and leading to the collapse of one or both wythes. In the presence of excessive openings, the wall becomes weak against the tearing action caused by the earthquake forces that are parallel to the length of the wall. This results in diagonal cracking of varying severity in the walls. Further, due to poor connection between the walls and the floor/roof, it results in failure of the floor diaphragm and affects the overall stability of the structure. Hence, while choosing to construct a building with stone masonry, the owner must make sure that the measures required to counter these weaknesses are taken during the construction so that in the event of a potentially destructive earthquake, the structure can withstand its impact without suffering much damage. Therefore, in the present study, an overview is presented for different failure modes of stone masonry houses, seismic safety provisions for the new construction considering the regulations for the wall, floor/roof construction and the provision of horizontal and vertical reinforcement to ensure good seismic performance against the future earthquakes.

Abstract

Evaluating the risk level of adversarial images is essential for safely deploying face authentication models in the real world. Popular approaches for physical-world attacks, such as print or replay attacks, suffer from some limitations, like including physical and geometrical artifacts. Recently adversarial attacks have gained attraction, which try to digitally deceive the learning strategy of a recognition system using slight modifications to the captured image. While most previous research assumes that the adversarial image could be digitally fed into the authentication systems, this is not always the case for systems deployed in the real world. This paper demonstrates the vulnerability of face authentication systems to adversarial images in physical world scenarios. We propose AdvGen, an automated Generative Adversarial Network, to simulate print and replay attacks and generate adversarial images that can fool state-of-the-art PADs in a physical domain attack setting. Using this attack strategy, the attack success rate goes up to 82.01%. We test AdvGen extensively on four datasets and ten state-of-the-art PADs. We also demonstrate the effectiveness of our attack by conducting experiments in a realistic, physical environment.

Abstract

A system for generating high-resolution de-warped omni-directional stereo image from captured omni-directional stereo image by correcting optical distortions using projection patterns is provided. The system includes a projection pattern capturing arrangement, a projector or a display, and a de-warping server. The projection pattern capturing arrangement includes one or more omnidirectional cameras to capture projection patterns from the captured omni-directional stereo image from each omni-directional stereo camera. The projector or the display displays the projection patterns. The de-warping server obtain the projection patterns and processes the projection patterns to generate high resolution de-warped omni-directional stereo image by correcting optical distortions in the captured omni-directional stereo image and mapping the captured omni-directional stereo image and the high resolution de-warped omni-directional stereo image.

Abstract

Text and signs around roads provide crucial information for drivers, vital for safe navigation and situational awareness. Scene text recognition in motion is a challenging problem, while textual cues typically appear for a short time span, and early detection at a distance is necessary. Systems that exploit such information to assist the driver should not only extract and incorporate visual and textual cues from the video stream but also reason over time. To address this issue, we introduce RoadTextVQA, a new dataset for the task of video question answering (VideoQA) in the context of driver assistance. RoadTextVQA consists of 3,222 driving videos collected from multiple countries, annotated with 10,500 questions, all based on text or road signs present in the driving videos. We assess the performance of state-of-the-art video question answering models on our RoadTextVQA dataset, highlighting the significant potential for improvement in this domain and the usefulness of the dataset in advancing research on in-vehicle support systems and text-aware multimodal question answering.

Abstract

Efficiently and accurately rendering hair accounting for multiple scattering is a challenging open problem. Path tracing in hair takes long to converge while other techniques are either too approximate while still being computationally expensive or make assumptions about the scene. We present a technique to infer the higher order scattering in hair in constant time within the path tracing framework, while achieving better computational efficiency. Our method makes no assumptions about the scene and provides control over the renderer’s bias & speedup. We achieve this by training a small multilayer perceptron (MLP) to learn the higher-order radiance online, while rendering progresses. We describe how to robustly train this network and thoroughly analyze our resulting renderer’s characteristics. We evaluate our method on various hairstyles and lighting conditions. We also compare our method against a recent learning based & a traditional real-time hair rendering method and demonstrate better quantitative & qualitative results. Our method achieves a significant improvement in speed with respect to path tracing, achieving a run-time reduction of 40% −70% while only introducing a small amount of bias.

Abstract

Significant progress has been made in speaker dependent Lip-to-Speech synthesis, which aims to generate speech from silent videos of talking faces. Current state-of-the-art approaches primarily employ non-autoregressive sequence-to-sequence architectures to directly predict mel-spectrograms or audio waveforms from lip representations. We hypothesize that the direct mel-prediction hampers training/model efficiency due to the entanglement of speech content with ambient information and speaker characteristics. To this end, we propose RobustL2S, a modularized framework for Lip-to-Speech synthesis. First, a non-autoregressive sequence-to-sequence model maps self-supervised visual features to a representation of disentangled speech content. A vocoder then converts the speech features into raw waveforms. Extensive evaluations confirm the effectiveness of our setup, achieving state-of-the-art performance on the unconstrained Lip2Wav dataset and the constrained GRID and TCD-TIMIT datasets. Speech samples from RobustL2S

Abstract

Traffic flow prediction has been regarded as a critical problem in intelligent transportation systems. An accurate prediction can help mitigate congestion and other societal problems while facilitating safer, cost and time-efficient travel. However, this requires the prediction algorithm to consider several complex characteristics of traffic flow data. These complex characteristics are an amalgamation of the spatial, temporal and periodic features exhibited by traffic flow data. To extract and leverage these features for traffic flow prediction, several hybrid deep learning models have been developed recently; however, there are still some challenges to determine the optimal architecture considering both spatial and temporal features. In this work, we perform an extensive comparison of hybrid deep learning models with and without periodicity to understand the prediction accuracy of these popular models. We propose an optimization framework that unifies genetic algorithm (GA) embedded optimization with prediction models in order to derive optimized deep learning architectures for hybrid traffic flow prediction exploring 2D spatial and temporal information. The framework enables the improvement of prediction performance and eliminates the handtuning process. An improved temporal convolutional network (TCN) architecture is derived using the GA driven optimization, which achieves superior traffic flow prediction accuracy compared to all other existing hybrid deep learning models on the freeway and urban traffic data from the PeMS traffic data set. We also evaluate the performance of the derived hybrid deep learning algorithms on the Raspberry PI embedded platform. Index Terms—Traffic flow prediction, Hybrid deep learning models, Temporal convolutional networks