Conference Papers
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506
Browse
7 results
Search Results
Item Image Colorization Using GANs and Perceptual Loss(Institute of Electrical and Electronics Engineers Inc., 2020) Sankar, R.; Nair, A.; Abhinav, P.; Mothukuri, S.K.P.; Koolagudi, S.G.Image colorization is of great use for several applications, such as the restoration of old images, as well as enabling the storage of grayscale images, which take up less space, which can later be colorized. But this problem is hard since there exist many possible color combinations for a particular grayscale image. Recent developments have aimed to solve this problem using deep learning. But, for achieving good performance, they require highly processed inputs, along with additional elements, such as semantic maps. In this paper, an attempt has been made for generalizing the procedure of colorization using a conditional Deep Convolutional Generative Adversarial Network (DCGAN) by adding "Perceptual Loss". The network is trained over the CIFAR-100 dataset. The results of the proposed generative model with perceptual loss are compared with the existing state-of-the-art systems normal GAN model and U-Net Convolutional model. © 2020 IEEE.Item Indian stock market prediction using deep learning(Institute of Electrical and Electronics Engineers Inc., 2020) Maiti, A.; Shetty D, P.In this paper, we predict the stock prices of five companies listed on India's National Stock Exchange (NSE) using two models- the Long Short Term Memory (LSTM) model and the Generative Adversarial Network (GAN) model with LSTM as the generator and a simple dense neural network as the discriminant. Both models take the online published historical stock-price data as input and produce the prediction of the closing price for the next trading day. To emulate the thought process of a real trader, our implementation applies the technique of rolling segmentation for the partition of training and testing dataset to examine the effect of different interval partitions on the prediction performance. © 2020 IEEE.Item Semantic Segmentation on Low Resolution Cytology Images of Pleural and Peritoneal Effusion(Institute of Electrical and Electronics Engineers Inc., 2022) Aboobacker, S.; Verma, A.; Vijayasenan, D.; Sumam David, S.; Suresh, P.K.; Sreeram, S.Automation in the detection of malignancy in effusion cytology helps to save time and workload for cytopathologists. Cytopathologists typically consider a low-resolution image to identify the malignant regions. The identified regions are scanned at a higher resolution to confirm malignancy by investigating the cell level behaviour. Scanning and processing time can be saved by zooming only the identified malignant regions instead of entire low-resolution images. This work predicts malignancy in cytology images at a very low resolution (4X). Annotation of cytology images at a very low resolution is challenging due to the blurring of features such as nuclei and texture. We address this issue by upsampling the very low-resolution images using adversarial training. This work develops a semantic segmentation model trained on 10X images and reuse the network to utilize the 4X images. The prediction results of low resolution images improved by 15% in average F-score for adversarial based upsampling compared to a bicubic filter. The high resolution model gives a 95% average F-score for high resolution images. Also, the sub-area of the whole slide that requires to be scanned at high magnification is reduced by approximately 61% while using adversarial based upsampling compared to a bicubic filter. © 2022 IEEE.Item A Study on Depth Estimation from Single Image Using Neural Networks(Institute of Electrical and Electronics Engineers Inc., 2022) Shree, R.; Madagaonkar, S.B.; Singh, M.; Chandra, M.T.A.; Rathnamma, M.V.; Venkataramana, V.; Chandrasekaran, K.Depth estimation is fundamental in upcoming technology advancements like scene understanding, robot vision, intelligent driver assistance systems, and many new technologies. Estimating the depth of objects from a viewport can be achieved using various mathematical, geometrical, and stereo concepts, but the process is unaffordable and erroneous. Depth estimation from a single can be accurately done using neural networks. Although this is a challenging task, researchers around the globe have published various works. The works include different neural network standards like CNN, GANs, Encoder-Decoder. The paper analyses and examines famous works in this field of study. Later in the paper, a comparative survey of depth estimation approaches using neural networks is done. © 2022 IEEE.Item Generalizing a Secure Framework for Domain Transfer Network for Face Anti-spoofing(Springer, 2023) Chandavarkar, B.R.; Rana, A.; Ketkar, M.M.; Pai, P.G.An essential field in cyber-security is the technology behind the authentication of users. In the contemporary era, alphanumeric passwords have been the primary tool used. A password is easy to understand but theoretically complex to brute force and is very easy to store in large databases since, in essence, they are an array of characters. In practical use, however, passwords tend to be very ineffective as the attributes that make for “strong†passwords are generally really abstract and random. However, human memories tend to remember constructed language, which can be intelligently guessed with sufficient information regarding the person. A widely accepted remedy to this problem is to replace passwords with face recognition software which involves minimal effort for the user and, though not completely immune, is quite challenging to misuse. Misuse is often done by spoofing faces. Spoofing faces means presenting a 2D or even a 3D copy of a face to the camera, pretending it is the real face. Many discovered spoofing methods are taken into consideration and protected against in most software. Even though various methods have been suggested for anti-spoofing, there are challenges that these methods go through, e.g., washed-out images, lack of colour, poor illumination. Our proposal involves the use of OpenCV and deep learning to accurately colourize the image pre-processing and put it through the GAN framework (Wang et al., From RGB to Depth: Domain Transfer Network for Face Anti-Spoofing (2021). https://ieeexplore.ieee.org/document/9507460. Accessed 07 Feb 2022). Furthermore, to secure the image to prevent ill use, the final generated image pair will be encrypted so that only the user can access it. The aim is to combine the power of re-colourization with the GAN framework’s strong anti-spoofing capabilities and make the whole framework secure for the user. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.Item Optimizing Super-Resolution Generative Adversarial Networks(Springer Science and Business Media Deutschland GmbH, 2023) Jain, V.; Annappa, B.; Dodia, S.Image super-resolution is an ill-posed problem because many possible high-resolution solutions exist for a single low resolution (LR) image. There are traditional methods to solve this problem, they are fast and straightforward, but they fail when the scale factor is high or there is noise in the data. With the development of machine learning algorithms, their application in this field is studied, and they perform better than traditional methods. Many Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) have been developed for this problem. The Super-Resolution Generative Adversarial Networks (SRGAN) have proved to be significant in this area. Although the SRGAN produces good results with 4 upscaling, it has some shortcomings. This paper proposes an improved version of SRGAN with reduced computational complexity and training time. The proposed model achieved an PPSNR of 29.72 and SSIM value of 0.86. The proposed work outperforms most of the recently developed systems. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.Item An Efficient AI and IoT Enabled System for Human Activity Monitoring and Fall Detection(Institute of Electrical and Electronics Engineers Inc., 2024) Verma, N.; Mundody, S.; Guddeti, R.M.R.Falls present a significant health risk, particularly among the elderly, necessitating reliable wearable fall detection systems. This paper introduces an advanced AI-powered system that integrates Generative Adversarial Networks (GANs) for synthetic data augmentation and Convolutional Neural Networks (CNNs) for robust fall detection and daily activity recognition. The primary challenge in developing effective fall detection systems lies in the scarcity and diversity of real-world fall data. This paper addresses this challenge innovatively by employing a GAN trained on datasets of authentic fall events to generate synthetic data. This augmentation strategy significantly expands the training dataset, enhancing the model's capacity to generalize across various fall scenarios and daily activities. The system leverages a specialized 1D CNN architecture designed for processing accelerometer and gyroscope readings obtained from wearable devices, enabling precise feature extraction to distinguish subtle differences between falls and routine movements. The evaluation results demonstrate a notable advancement by achieving a superior accuracy of 99 % for fall detection while minimizing false positives. The developed CNN model can also classify 15 kinds of falls and 19 types of daily life activities. © 2024 IEEE.