Manjunath D

Manjunath D

Robotic perception & Autonomous driving

Greetings, Universe! I’m Manjunath

I’m currently working as Research Assistant at the Artifitial Intelligence and Robotics Lab (AIRL) in the Department of Aerospace Engineering at the Indian Institute of Science (IISc).
I work under the guidance of Prof. Suresh Sundaram on research projects in the domain of computer vision that is applied to ground and aerial autonomous vehicles. I completed my Bachelors in Information Science and Engineering in 2022 from Atria Institute of Technology, Bengaluru.

News / Updates:

Apr 2025          Two papers accpted at CVPR-W 2025 (PBVS & DG-EBF)

Jan 2023          Joined Artifitial Intelligence and Robotics Lab (AIRL) under Prof. Suresh Sundaram

Aug 2022          Graduated B.Tech with First Class with Honors from Atria Institute of Technology

Jan 2022          Joined Energy and Wetlands Research Group under Prof. T V Ramachandra

Publications:

SAGA: Semantic-Aware Gray color Augmentation for Visible-to-Thermal Domain Adaptation across Multi-View Drone and Ground-Based Vision Systems
*Manjunath D*, Aniruddh Sikdar, Prajwal Gurunath, Sumanth Udupa, Suresh Sundaram
Accpted at CVPR-W (PBVS) (Oral presentation)

VISTA-CLIP: Visual Incremental Self-Tuned Adaptation for Efficient Continual Panoptic Segmentation
*Manjunath D*, Shrikar Madhu, Aniruddh Sikdar, Suresh Sundaram
Accpted at CVPR-W (DGEBF)

Research Projects:

  • Continual learning using Vision Language Models (VLM) and visual prompting techniques with knowledge distillation for autonomous vehicles for Panoptic Segmentation task
  • Multi-modal unsupervised RGB to Thermal Domain adaptation (Un co-registered images) for Autonomous Ground Vehicles using self supervised by leveraging the advantage of pretext tasks.
  • Slant angle object detection to tackle the challenges of occlusion, scale variations and domain shifts due to weather, geography, sensor, etc in drone imagery
  • Improving accuracy for object detection on Synthetic Aperture Radar (SAR) dataset.

    • Presentations and Talks:

    Paper Presentation:
  • Oral Presentation of SAGA: Semantic-Aware Gray color Augmentation for Visible-to-Thermal Domain Adaptation across Multi-View Drone and Ground-Based Vision Systems" at CVPR-W 2025, in Nashville TN, US
  • Poster Presentation of "VISTA-CLIP: Visual Incremental Self-Tuned Adaptation for Efficient Continual Panoptic Segmentation" at CVPR-W 2025, in Nashville TN, US
    • Talks:
  • Delivered a talk on Target Recognition in aerial imagery, Faculty Development Program for "Drone Perception", PES University, Electronic City Campus, Bangalore, Karnataka, India
  • Delivered a talk on Deep leraning for autonomous navigation, Industrial workshop for Ashok Leyland on "Autonomous Navigation", Department of Aerospace Engineering (AE), IISc, Bangalore, Karnataka, India
  • Delivered a talk on Deep Learning for Drone Applications, Faculty Development Program for "Drone Applications", Department of AE, IISc,. Bangalore, Karnataka, India
  • Delivered a talk on AI perception for Aerial Robotics, Faculty Development Program for "Unmanned Aerial Vehicles Technology and its Application", Department of AE, IISc,. Bangalore, Karnataka, India