Satyajit Tourani At present I am working at TCS-Research as a Predoctoral Research Fellow under Dr. Brojeshwar Bhowmick in the VCEI group. Here, I am working on the problem of Social Robot Navigation. At IIIT Hyderabad, I did my master's degree in research. Under Prof. Madhava Krishna's supervision at the Robotics Research Center. During my master's programme, I focused mostly on the topic of visual place recognition in both indoor and outdoor settings. I've published my work at prestigious robotics and vision conferences like IROS, ICRA and VISAPP. Prior to joining IIIT Hyderabad, I worked at Amazon as an SDE intern. I was a member of the INGC (Indian Giftcard) team. I was in charge of building a system which involved the end consumer to receive an SMS message anytime he performed a giftcard-related action (Buy, Redeem, Expiry etc). My other interests lie in competitive programming and playing fps games. Email  /  GitHub  /  CV  /  Google Scholar  /  LinkedIn

Abstract-coming soon.

In this paper, we present a novel framework that combines the learning of invariant descriptors through data augmentation and orthographic viewpoint projection. We propose rotation-robust local descriptors, learnt through training data augmentation based on rotation homographies, and a correspondence ensemble technique that combines vanilla feature correspondences with those obtained through rotation-robust features.

Significant recent advances have been made in Visual Place Recognition (VPR), feature correspondence and localization due to deep-learning-based methods. However, existing approaches tend to address, partially or fully, only one of two key challenges: viewpoint change and perceptual aliasing. In this paper, we present novel research that simultaneously addresses both challenges by combining deep-learnt features with geometric transformations based on domain knowledge about navigation on a ground-plane, without specialized hardware (e.g. downwards facing cameras, etc.).

In this work We showcase a topological mapping framework for a challenging indoor warehouse setting. At the most abstract level, the warehouse is represented as a Topological Graph where the nodes of the graph represent a particular warehouse topological construct (e.g. rackspace, corridor) and the edges denote the existence of a path between two neighbouring nodes or topologies. At the intermediate level, the map is represented as a Manhattan Graph where the nodes and edges are characterized by Manhattan properties and as a Pose Graph at the lower-most level of detail. The topological constructs are learned via a Deep Convolutional Network while the relational properties between topological instances are learnt via a Siamese-style Neural Network. In the paper, we show that maintaining abstractions such as Topological Graph and Manhattan Graph help in recovering an accurate Pose Graph starting from a highly erroneous and unoptimized Pose Graph. We show how this is achieved by embedding topological and Manhattan relations as well as Manhattan Graph aided loop closure relations as constraints in the backend Pose Graph optimization framework. The recovery of near ground-truth Pose Graph on real-world indoor warehouse scenes vindicate the efficacy of the proposed framework.