Ananya Vivek Kulkarni

I am a MS by Research (CSE) student at IIIT Hyderabad, advised by Prof. C. V. Jawahar. I am also an AI/ML Intern at iHub-Data, IIIT Hyderabad. My work focuses on computer vision, Optimization, Image-Video Denoising, and knowledge distillation for adverse weather removal and real-time edge deployment. Previously, I worked as a Research Intern at SAC, ISRO on radiative transfer emulation and at DIAT, DRDO on deep learning–based leak detection.

I received my B.Tech in Information Technology from Dharmsinh Desai University.

I have received a Best Paper Award (Springer CCIS, 2023), have publications in OTCON 2025, and papers accepted at ERCICAM 2025 and ICVGIP 2025, along with a patent on a real-time text-to-Braille conversion system.

Email  /  CV  /  Bio  /  Scholar  /  Github  /  LinkedIn

profile photo

Research

My research focuses on image and video restoration, particularly denoising and adverse weather removal, along with efficient model design through knowledge distillation. I aim to improve visual perception systems for challenging real-world environments.

Areas of Interest: Image and Video Restoration, Video Processing, Knowledge Distillation, Object Detection, Computer Vision.
Some papers are highlighted.

Enhancing Driving Visibility via Semantic-Guided Knowledge Distillation for Adverse Weather Removal
H. S. Mukkamala, Ananya Vivek Kulkarni, S. Gangisetty, V. G. Yalla, C. V. Jawahar
ICVGIP, 2025
code

We propose a semantic-guided knowledge distillation framework that improves visibility restoration under adverse weather while enabling efficient real-time deployment on edge devices.

Improving Vehicle Visibility in Fog Environment
Ananya Vivek Kulkarni, Zankhana Barad, Harshadkumar B. Prajapati
OTCON, 2025
paper

This work combines image enhancement techniques (DCP, CLAHE) with deep learning-based detection and tracking to improve vehicle visibility and robustness in foggy driving scenarios.

Gender Agreement in Indo-Aryan Languages Using Rule-Based Parsing and Finite Automata
Ananya Vivek Kulkarni, Srushti Pednekar, Deepak Vegda
ERCICAM, 2025
preprint

We present a rule-based syntactic parsing framework using finite automata to model gender agreement in Indo-Aryan languages, combining linguistic theory with computational efficiency.

Raspberry Pi-Driven Affordable Image-to-Braille Converter for Visually Impaired Users
Ananya Vivek Kulkarni, Maitri Shah, Nivedita Thakur, Srushti Pednekar, Viral H. Shah
Springer CCIS, 2024 (Best Paper Award)
paper

We propose a low-cost, real-time image-to-Braille conversion system using OCR and Raspberry Pi, designed to improve accessibility to educational content for visually impaired users.

Women Academicians in Engineering Colleges in Gujarat, India: A Survey of the Current Scenario
Srushti Pednekar, Ananya Vivek Kulkarni, Maitri Shah, Nikita Desai
Conference Paper, 2024
preprint

This survey analyzes participation, challenges, and representation of women academicians in engineering institutions across Gujarat, highlighting structural and institutional trends.

Miscellanea

Honors
Best Paper Award, ASCIS 2023 (Springer CCIS)
Student Startup Innovation Policy (SSIP), Govt. of Gujarat (funded student project)
Top-10 Finalist, DevHeat Beta Hackathon (student category)

Presentations
Paper presentation at ERCICAM 2025 (rule-based parsing for Indo-Aryan languages)
Conference presentation at OTCON 2025 (vehicle visibility in foggy environments)
Best Paper presentation at ASCIS 2023
Technical session on Git Fundamentals for TRAIL Team, INAI Mobility, CVIT, IIIT Hyderabad (Feb 2026) [slides]
Technical session on Introduction to Linux for TRAIL Team, INAI Mobility, CVIT, IIIT Hyderabad (Feb 2026) [slides]
Technical session on Linux Command Line Fundamentals for TRAIL Team, INAI Mobility, CVIT, IIIT Hyderabad (Feb 2026) [slides]

Patents
A System and Method for Real-Time Text-to-Braille Converter
Indian Patent Application No. 202421066714, published Dec 2024

Beyond Research
Classical Bharatanatyam dancer (trained for 3 years, distinction)
Contributor to open-source climate ML projects (OpenClimateFix)
Interested in edge AI systems, low-visibility perception, and 3D scene understanding