Research Projects

Vaishnavi Khindkar, Vineeth N Balasubramanian, Chetan Arora, Anbumani Subramanian, C V Jawahar

IEEE International Conference on Intelligent Robots and System (IROS) 2024

Paper / Talk / Dataset and Code / Graphical_Abstract / Poster / Publication Link

With the increased importance of autonomous navigation systems has come an increasing need to protect the safety of Vulnerable Road Users (VRUs) such as pedestrians. Predicting pedestrian intent is one such challenging task, where prior work predicts the binary cross/no-cross intention with a fusion of visual and motion features. However, there has been no effort so far to hedge such predictions with human- understandable reasons. We address this issue by introducing a novel problem setting of exploring the intuitive reasoning behind a pedestrian’s intent. In particular, we show that predicting the ‘WHY’ can be very useful in understand- ing the ‘WHAT’. To this end, we propose a novel, reason- enriched PIE++ dataset consisting of multi-label textual expla- nations/reasons for pedestrian intent. We also introduce a novel multi-task learning framework called MINDREAD, which leverages a cross-modal representation learning framework for predicting pedestrian intent as well as the reason behind the intent. 

Vaishnavi Khindkar, Vineeth N Balasubramanian, Chetan Arora, Anbumani Subramanian, Rohit Saluja, C V Jawahar

IEEE Winter Conference on Applications of Computer Vision, WACV 2022

Paper / Talk / Code / Poster / Supplementary / Leaderboard / Publication Link

Adaptive object detection remains challenging due to visual diversity in background scenes and intricate combinations of objects. Motivated by structural importance, we aim to attend prominent instance-specific regions, overcoming the feature misalignment issue. We propose a novel resIduaL seLf-attentive featUre alignMEnt ( ILLUME ) method for adaptive object detection. ILLUME comprises Self-Attention Feature Map (SAFM) module that enhances structural attention to object-related regions and thereby generates domain invariant features. Our approach significantly reduces the domain distance with the improved feature alignment of the instances.

Vaishnavi Khindkar, Janhavi Khindkar

Paper / Code / Training Video / Working Video

We propose an efficient and generative augmentation approach to solve the inadequacy concern of underwater debris data for visual detection. We use cycleGAN as a data augmentation technique to convert openly available, abundant data of terrestrial plastic to underwater-style images. Prior works just focus on augmenting or enhancing existing data, which moreover adds bias to the dataset. Compared to our technique, which devises variation, transforming additional in-air plastic data to the marine background.

Vaishnavi Khindkar, Aishwarya Koppella, Ashwini Adhau, Sharayu Pardeshi, Mahalaxmi Reddy

Paper / Code

This project provides controlling and monitoring of home appliances as well as provides security from unknown persons. We proposed a system for Smart Home Automation technique. To design this system, we used a Raspberry Pi module and Computer Vision techniques, OpenCV and image processing algorithms.