Madhu Vankadari

I am a 2nd year DPhil student at the University of Oxford, working under the supervision of Prof. Niki Trigoni and Prof. Andrew Markhem.

My undergraduate degree is in Mechanical Engineering at Rajiv Gandhi University of Knowledge Technologies and I never did my Masters Degree.

CV   /   Google Scholar  /  LinkedIn  /  Github /  Email 

oxford
tcs
tcs
profile photo
Research Interests

I worked with many robots such as manipulators, ground and Aerail Vechicles. I aim to make these robots a little bit safer than today. Currently, I am working on large-scale localization and mapping problems which has potential applications in Autonomous Driving, Augmented and Virtual Reality areas.

Skills: Python | Tensorflow | PyTorch | ROS | Gazebo

Published Venues: ECCV(1) ✔ IJCAI(1) ✔ ICRA(3) ✔ IROS(3) ✔ CORL(2) ✔

Publications

2022 - 2020 | 2020 - 2018 | 2018 - before


new-paper
When the Sun Goes Down: Repairing Photometric Losses for All-Day Depth Estimation
Madhu Vankadari, Stuart Golodetz, Sourav Garg, Sangyun Shin, Andrew Markham, Niki Trigoni

In this paper, we show how to use a combination of three techniques to allow the existing photometric losses to work for both day and nighttime images

CORL 2022| Abstract | Paper | Code


new-paper
Real-Time Hybrid Mapping of Populated Indoor Scenes using a Low-Cost Monocular UAV
Stuart Golodetz, Madhu Vankadari, Aluna Everitt, Sangyun Shin, Andrew Markham, Niki Trigoni

We present what is thus, to our knowledge, the first system to perform simultaneous mapping and multi-person 3D human pose estimation from a monocular camera mounted on a single UAV.

IROS, 2022 | Abstract| Paper

Attentive One-Shot Meta-Imitation Learning From Visual Demonstration
Vishal Bhutani, Madhu Vankadari, Anima Majumder, Ardhendu Behera Swagat Kumar,

An attempt is made to solve this problem in this paper by proposing a deep meta-imitation learning framework comprising of an attentive-embedding net- work and a control network, capable of learning a new task in an end-to-end manner while requiring only one or a few visual demonstrations

ICRA, 2022 | Abstract

SeqMatchNet: Contrastive Learning with Sequence Matching for Place Recognition & Relocalization
Sourav Garg, Madhu Vankadari, Michael Milford

In this work, for the first time, we bridge the gap between single image representation learning and sequence matching through" SeqMatchNet" which transforms the single image descriptors such that they become more responsive to the sequence matching metric

CORL, 2021| Abstract | Paper | Code

Unsupervised Monocular Depth Estimation for Night-time Images using Adversarial Domain Feature Adaptation
Madhu Vankadari, Sourav Garg, Swagat Kumar, Anima Majumder, Ardhendu Behera

In this paper, we look into the problem of estimating per-pixel depth map for monocular night images.

ECCV, 2020| Abstract | Paper

Unsupervised Depth and Confidence Prediction from Monocular Image using Bayesian Inference
Vishal Bhutani, Madhu Vankadari, Omprakash Jha, Anima Majumder, Samrat Dutta, Swagat Kumar

In this paper, we propose an unsupervised deep learning framework with Bayesian inference for improving the accuracy of per-pixel depth prediction from monocular RGB images.

IROS, 2020 | Abstract| Paper

Attentive Task-Net: Self Supervised Task-Attention Network for Imitation Learning using Video Demonstration
Kartik Ramachandruni, Madhu Vankadari, Anima Majumder, Samrat Dutta, Swagat Kumar

In this paper we present a new depth and temporal-aware visual place recognition system that solves the opposing viewpoint, extreme appearance-change visual place recognition problem.

ICRA, 2020 | Abstract | Video | Presentation