Akash Kumbar
Akash Kumbar

I build perception and autonomy systems for robots operating in the real world, from 3D reconstruction and scene understanding to sensor calibration and navigation.

I’m Akash Kumbar, an M.S. by Research student and Graduate Student Researcher at the Robotics Research Center, IIIT Hyderabad, advised by Prof. Madhava Krishna.

My research has been published at ICCV 2025 and the ICCV 2023 Workshops, with work accepted to IROS 2026. I also review papers for ICVGIP 2026 and an IROS 2026 workshop.

Previously, I studied Electronics and Communication Engineering at KLE Technological University and researched point-cloud upsampling at the Center of Excellence in Visual Intelligence.

Outside research, I train in mixed martial arts and hold a black belt in karate. I also enjoy running and am currently training for a half marathon.

Updates

Research

* indicates equal contribution

Beyond Frontiers preview Beyond Frontiers: Scene-Anomaly Guided Autonomous Exploration

Akash Kumbar, Abhinav Raundhal, Madhava Krishna
Proceedings of the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)
Project Page  •   ArXiv

Dynamic Urban Scene Rendering preview Leveraging 2D Priors and SDF Guidance for Dynamic Urban Scene Rendering

Siddharth Tourani, Jayaram Reddy, Akash Kumbar, Satyajit Tourani, Nishant Goyal, Madhava Krishna, N Dinesh Reddy, Muhammad Haris Khan
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2025
Project Page  •   Proceedings  •   ArXiv

TP-NoDe preview TP-NoDe: Topology-Aware Progressive Noising and Denoising of Point Clouds Towards Upsampling

Akash Kumbar*, Tejas Anvekar*, Tulasi Amitha Vikrama, Ramesh Ashok Tabib, Uma Mudenagudi
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2023
Proceedings

ASUR3D preview ASUR3D: Arbitrary Scale Upsampling and Refinement of 3D Point Clouds Using Local Occupancy Fields

Akash Kumbar, Tejas Anvekar, Ramesh Ashok Tabib, Uma Mudenagudi
Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2023
Proceedings

Teaching & Service

Teaching Assistant, Mobile Robotics, IIIT Hyderabad, Monsoon 2026

Mentoring students, supporting course instruction, designing projects and assignments, and grading coursework.

Instructor, RRC Summer School, IIIT Hyderabad, 2025–2026

Taught deep learning for 3D vision, neural representations, and explicit scene representations, including 3D Gaussian Splatting. 2026 lecture

Reviewer, ICVGIP 2026

Reviewing submissions for ICVGIP 2026.

Reviewer, IROS 2026 Workshop(The name of the workshop will be made public after proceedings)

Reviewing submissions for a workshop at IROS 2026.

Projects

A selection of my work in computer vision, deep learning, and robotics.

LiDAR–Camera Calibration Robust target-based calibration of six cameras against a 64-channel LiDAR View project → Dynamic Obstacle Detection and Avoidance Multi-modal perception and reactive planning for collision-free Husky navigation View project → ARROU adversarially robust rank-one unlearning pipeline ARROU: Adversarially Robust Rank-One Unlearning Targeted LLM unlearning through latent adversarial training and localized rank-one model editing View project → MAST3R-SLAM Camera-pose extraction and 3D scene reconstruction from point clouds and images View project → Pothole and speed breaker detection result Camera-Based Vehicle Functions Completed as a part of Bosch's PRIXEL program View project →
View all projects →

Template borrowed from Cheng Chi's personal page