Applied Data Scientist building scalable, privacy-compliant AI systems for autonomous driving. Specialized in LL-
MOps,

Omprakash Vishwakarma  • Peerlist

Aim is to develop a software which minimizes physical interaction between humans and computers by computer vision that recognizes hand gestures to perform the task. Worked on Computer Vision and Convolutional Neural Network

Hand gesture Recognition and Intelligent Character recogniti

Nowadays, OOT platforms like Disney + Hotstar, SONY Liv, and others perform live-streaming of various events. Video quality is a significant concern in countries like India, and network strength is relatively weak. So the proposed work was in the same direction to develop a machine learning model that takes a low-resolution input image up-scaled to the high-resolution space; based on Convolutional Neural Network.

Image Super-Resolution using CNN

Legal Judgement is the task of predicting the outcome of a court judgment, given the textual description concerning the case. The data contains cases of European Court for Human Rights (ECHR) violations. Our focus is on a Neural approach to predict the violation of human rights based on the case description. The previous approaches do not explore the explicit property of text data gathering information from their neighbourhood and reflecting similar structure as a graph shows in the same feature space. But, the dominant graph neural networks heavy relied on graph links, and while training such graph models, they lead to over-smoothening and suspended animation problems. A fair Graph-BERT approach is proposed to overcome these issues, which does not consider the meta-information of the case description but tries to capture information from some neighbours and eliminate the over-smoothening and suspended animation problems while predicting the verdicts.

Neural Jugement Prediction in English

Aim is to develop a model that takes two images as input and blend them together to form an input image such that the transformation looks like content image and the paint is similar to the style image. Worked on Computer Vision and Convolutional Neural Network

Neural Style Transfer

A literature survey on graph neural network approaches to overcome the limitations of RNN and CNN alone by covering the internal structure such as semantics and syntactic information of the text data. It is robust to heterogeneous data for text classification.

Text Classification using Graph Convolution Network

Social media platforms are meant to interact with people around them and build a good network but it is used for spreading misinformation and disinformation in recent years. This is a major issue in today's world. A novel approach is proposed to use a graph-based model for detection of the vulnerable nodes which can spread misinformation/ disinformation.

The impact of noise on the social media

Topological structures are quite similar when observed from a higher dimensional subspace. This can be viewed in sparse clusters. So, we implemented a proposed algorithm LARS-LASSO which combines statistical and topological criteria for model selection inspired by LASSO.

Omprakash Vishwakarma

Applied Data Scientist building scalable, privacy-compliant AI systems for autonomous driving. Specialized in LL- MOps,