Surbhi Agarwal(Indian Institute of Technology , Hyderabad, 2022)
Humans can continuously accumulate, develop, and transfer knowledge and skills
throughout their lifetimes, giving rise to continual learning principles. However, Isolated learning is currently the most used paradigm in ...
Vinay Prakash; C. Krishna Mohan(Indian Institute of Technology, Hyderabad, 2021)
In computer vision, image segmentation deals with the task of segregating the pixels of an
image into different groups or segments. The goal of image segmentation is to recognize patterns,
boundaries etc. in the image, ...
D V K M Rishab; C. Krishna Mohan(Indian Institute of Technology , Hyderabad, 2022)
Point Cloud Segmentation is the process of categorizing multiple sets of point clouds into homogeneous regions where each region has its relevance which means that all the points in one area will
have similar characteristics. ...
Priya Bhatia; Soumya Jana(Indian Institute of Technology , Hyderabad, 2022)
During the COVID-19 pandemic, the world faced a global health crisis because of its contagious nature.
Even a person who is asymptomatic and has COVID-19 can spread the virus. At that point, an enormous volume of lung ...
Shaik Zeeshan Ali; J. Saketha Nath(Indian Institute of Technology , Hyderabad, 2022)
We consider the task of aligning two sets of high-dimensional embeddings using Optimal Transport(OT) where we have indirect supervised data. Consider three sets of embeddings(languages),
source, bridge, and target embeddings ...
Saransh Bali; Srijith P.K(Indian Institute of Technology , Hyderabad, 2022)
Modeling when and where discrete events (localized in continuous space and time) will occur is a
challenging and essential problem. This modeling can include, for example, earthquakes and after shocks, check-ins in a city, ...
Matish Singh Tanwar; Sathya Peri(Indian Institute of Technology , Hyderabad, 2022)
We have solved two optimization problems using Genetic Algorithm, one is on Image Reconstruction(given an image we have make a copy which looks most similar to it) and other is based on civil
engineering domain problem ...
Background & objectives: To handle the current COVID-19 pandemic in India, multiple strategies have been applied and implemented to slow down the virus transmission. These included clinical management of active cases, rapid ...
Tammana, Praveen(Association for Computing Machinery, Inc, 2021-08-23)
Programmable data planes provide exciting opportunities to realize fast, accurate, and data-driven control-loop decisions. Many data plane systems have been proposed for handling network dynamics (e.g., congestion, failures) ...
The thesis presents a novel approach to signal processing and face recognition based on 60 GHz
mmWave RADAR imaging. Machine learning algorithms such as Convolutional Auto-Encoder and
Random Forest algorithm are employed ...
A R, Sreeragh; C. Krishna Mohan(IIT HYDERABAD, 2021-08-07)
This Masters Thesis explores various state-of-the-art semantic segmentation models and ways to
improve the performance. Contextual information is crucial in semantic segmentation task. With
normal CNN (Convolutionsl ...
Probabilistic Generative Modeling is a significant field that is widely used for various applications
involving diverse domains. Its application ranges from everyday use by common-man such as chat bots to the scientific ...
Reddy Bhavanam, Akhil Kumar; Mohan, C. Krishna(IIT HYDERABAD, 2021-07)
According to the World Health Organization(WHO) report, 2021 [1], there are globally atleast 2.2 bil lion people who have a vision impairment or blindness.According to the medical journal ‘The Lancet
Global Health’ [2], ...
There is major requirement especially in the defense sector to be able to detect and identify hostiles
during both the day and night as well. The night poses a more daunting task for detecting the
enemy. If we get a live ...
Neural Networks which are highly useful models for state of the art models for many artificial
intelligence tasks, suffer from brittleness i.e., very small imperceptible changes in input can cause
a serious change in ...
Online Video panoramic stitching finds applications in various domains like VR, Surveillance,
Navigation. For Feature based image stitching using Feature detection description has high inference
time ,which is not ideal ...
Given the necessary information required, predicting the target is a matter of architecture design
and fine-tuning. But what if all the necessary information is not present during inference? Consider
an example of ...
P Sylvan, Joseph; P. K, Srijith(IIT HYDERABAD, 2021-07)
With the vast modern advances of technology past a few decades, information such as audio, video
or text is transferred among individuals in the blink of an eye with the help of social networks such
as Twitter, Facebook, ...
Prakash, Vinay; Mohan, C Krishna(IIT HYDERABAD, 2021-06)
In computer vision, image segmentation deals with the task of segregating the pixels of an
image into different groups or segments. The goal of image segmentation is to recognize patterns,
boundaries etc. in the image, ...
Object detection which is a computer vision problem mainly deals with
identifying and locating the objects which belongs to different classes in
the image/video. Now a days the world is emerging with usage of new
technologies ...