CV

Education

• B.Tech, Electronics and Instrumentation Engineering, NIT Rourkela, 2025 → 8.65/10.0
• Class 12 (CBSE), ODM Public School, 2021 → 96.2%
• Class 10 (ICSE), St. Joseph’s High School, 2019 → 98%

Thesis

• Joint Haze and Sand-dust Image Restoration for Effective Object Detection and Classification
  • Undergraduate Thesis, National Institute of Technology Rourkela, 2025
  • Advisor: Dr. Sobhan Kanti Dhara

Work experience

• Perception and Recognition via Imaging and Sensing using Modern-AI (PRISM) Research Group, NIT Rourkela, 2024-2025: Undergraduate Research Assistant

  • Advisor: Dr. Sobhan Kanti Dhara
    • Developed an encoder-decoder based framework to restore image degradations caused by haze and sand-dust for improving object detection and classification performance.
    • Incorporated selective attention mechanisms to emphasize relevant features while reducing computation.
• Machine Intelligence and Bio-motion Lab, NIT Rourkela, 2024: Undergraduate Research Assistant
  • Advisor: Dr. Anup Nandy
    • Implemented a DRL-based navigation pipeline using the TD3 algorithm and laser scan inputs to enable decision-making in dynamic environments, and deployed it on an NVIDIA Jetson Nano, balancing inference speed with performance constraints.
    • Developed a SLAM-based indoor navigation and exploration framework, focusing on static environments using LiDAR data and spatial mapping techniques.
    • Designed a transformer-based architecture for speech emotion recognition, combining acoustic features (pitch, tone) with transcript-based textual cues to support emotionally aware human-robot communication in multimodal settings.
• Barclays, Summer 2024: Global Security Developer Intern
  • Manager: Atul Mishra
    • Enhanced system efficiency by reducing synchronization time by a factor of 8 through parallel processing and dynamic server retrieval using server aliases.
    • Automated job scheduling with Autosys integration, eliminating manual intervention and improving reliability.
• Computer Vision Lab, IIT Kharagpur, Summer 2023: Research Intern
  • Advisor: Prof. Aurobindo Routray
    • Developed a hybrid CNN–Transformer model for robust face detection and recognition in video sequences, with a focus on handling motion blur and partial occlusion.

Skills

Relevant Coursework

• Computer Vision
• Deep Learning
• Autonomous Navigation
• Object Oriented Programming
• Data Structures and Algorithms
• Digital Image Processing
• Machine Intelligence
• Probability and Statistics

Talks

• NIT Rourkela Institute Seminar on “Towards Single Sand-Dust Image Restoration via Vision Transformer with Multi-Scale Feature Aggregation”.

Journal Reviewing

• Circuits, Systems & Signal Processing - 2025

Conference Reviewing

• 6th IEEE India Council International Subsections Conference - 2025
• 4th International Conference on Computer Vision and Machine Intelligence - 2025

Honors and Awards

• Best UG Project Finalist, NIT Rourkela (2025)
• CVPR Travel grant (2025)
• Special Recognition at ICORT, DRDO (2025)
• Semi-finalist at eYantra Robotics Challenge, IIT Bombay (2023)
• Second Runners-up at Voyager Competition, IIT Kharagpur (2023)

Publications

Dust to Detail: Restoring Sand-dust Images with Frequency-Guided Attention and Multi-Scale Features

Adaptive Contextual Feature Fusion: Leveraging Human-Robot Interaction with Speech Emotion Recognition

Towards Single Sand-Dust Image Restoration via Vision Transformer with Multi-Scale Feature Aggregation

For a detailed CV/resume please email me at romalamishra10@gmail.com