Abhishek Potnis

Mr. Abhishek Potnis

Indian Institute of Technology Bombay, India
orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 July 2019
Verified
Conference Paper
Closed access
UBUjwala Bhangale
Ujwala Bhangale
SDSurya Durbha
Surya Durbha
APAbhishek Potnis
Abhishek Potnis
RSRajat Shinde
Rajat Shinde
Very High Resolution (VHR) remote sensing optical imagery is a huge source of information that can be utilized for earthquake damage detection and assessment. Time critical task such as performing the damage assessment, providing immediate delivery of relief assistance require immediate response; however, processing voluminous VHR imagery using highly accurate, but computationally expensive deep learning algorithms demands the High Performance Computing (HPC) power.To maximize the accuracy, deep convolution neural network (CNN) model is designed especially for the earthquake damage detection using remote sensing data and implemented using high performance GPU without compromising with the execution time. Geoeye1 VHR disaster images of the Haiti earthquake occurred in year 2010 is used for analysis. Proposed model provides good accuracy for damage detection; also significant execution speed is observed on GPU K80 High Performance Computing (HPC) platform.

Verified
Journal Article
Open access
PDF file
APAbhishek Potnis
Abhishek Potnis
SDSurya Durbha
Surya Durbha
RSRajat Shinde
Rajat Shinde
Earth Observation data possess tremendous potential in understanding the dynamics of our planet. We propose the Semantics-driven Remote Sensing Scene Understanding (Sem-RSSU) framework for rendering comprehensive grounded spatio-contextual scene descriptions for enhanced situational awareness. To minimize the semantic gap for remote-sensing-scene understanding, the framework puts forward the transformation of scenes by using semantic-web technologies to Remote Sensing Scene Knowledge Graphs (RSS-KGs). The knowledge-graph representation of scenes has been formalized through the development of a Remote Sensing Scene Ontology (RSSO)—a core ontology for an inclusive remote-sensing-scene data product. The RSS-KGs are enriched both spatially and contextually, using a deductive reasoner, by mining for implicit spatio-contextual relationships between land-cover classes in the scenes. The Sem-RSSU, at its core, constitutes novel Ontology-driven Spatio-Contextual Triple Aggregation and realization algorithms to transform KGs to render grounded natural language scene descriptions. Considering the significance of scene understanding for informed decision-making from remote sensing scenes during a flood, we selected it as a test scenario, to demonstrate the utility of this framework. In that regard, a contextual domain knowledge encompassing Flood Scene Ontology (FSO) has been developed. Extensive experimental evaluations show promising results, further validating the efficacy of this framework.

29 October 2024
Verified
Conference Paper
Open access
PDPhilipe Dias
Philipe Dias
ATAristeidis Tsaris
Aristeidis Tsaris
JBJordan Bowman
Jordan Bowman
APAbhishek Potnis
Abhishek Potnis
JAJacob Arndt
Jacob Arndt
OLORNL Lunga
ORNL Lunga
While the pretraining of Foundation Models (FMs) for remote sensing (RS) imagery is on the rise, models remain restricted to a few hundred million parameters. Scaling models to billions of parameters has been shown to yield unprecedented benefits including emergent abilities, but requires data scaling and computing resources typically not available outside industry R&D labs. In this work, we pair high-performance computing resources including Frontier supercomputer, America's first exascale system, and high-resolution optical RS data to pretrain billion-scale FMs. Our study assesses performance of different pretrained variants of vision Transformers across image classification, semantic segmentation and object detection benchmarks, which highlight the importance of data scaling for effective model scaling. Moreover, we discuss construction of a novel TIU pretraining dataset, model initialization, with data and pretrained models intended for public release. By discussing technical challenges and details often lacking in the related literature, this work is intended to offer best practices to the geospatial community toward efficient training and benchmarking of larger FMs.

1 July 2019
Verified
Conference Paper
Closed access
APAbhishek Potnis
Abhishek Potnis
RSRajat Shinde
Rajat Shinde
SDSurya S. Durbha
Surya S. Durbha
KKKuldeep R. Kurte
Kuldeep R. Kurte
Natural disasters such as floods, earthquakes, hurricanes, etc. have a huge impact on a society—causing destruction of life and property in their wake. During disasters such as flood, it is crucial to understand the dynamics of the situation as it occurs for effective response. In this paper, we address the problem of satellite image classification for urban floods using deep learning. We propose an encoder-decoder neural network based on the Efficient Residual Factorized Convnet(ERFNet), for multi-class segmentation of urban floods from multi-spectral satellite imagery. The ERFNet architecture capitalizes on skip connections and one dimensional convolutions to achieve the best possible trade-off between accuracy and efficiency. Since time is of essence during a disaster, the choice of the ERFNet architecture on a high performance computing (HPC) platform is apt. Satellite imagery from WorldView-2 of floods in Srinagar, India during September 2014 have been used for this study. The tool ‘markGT’ has been developed to assist end-to-end annotation of satellite imagery. The urban flood dataset used for this study has been generated using markGT. The proposed deep learning model over urban flood satellite imagery gives promising results on Nvidia Tesla K80 GPU. We envisage that the proposed model could be extended and improved for real-time classification of urban floods, thereby aiding disaster response personnel in making informed decisions.

5 November 2019
Verified
Conference Paper
Open access
PDF file
KKKuldeep Kurte
Kuldeep Kurte
APAbhishek Potnis
Abhishek Potnis
SDSurya Durbha
Surya Durbha
Extreme events such as urban floods are dynamic in nature, i.e. they evolve with time. The spatiotemporal analysis of such disastrous events is important for understanding the resiliency of an urban system during these events. Remote Sensing (RS) data is one of the crucial earth observation (EO) data sources that can facilitate such spatiotemporal analysis due to its wide spatial coverage and high temporal availability. In this paper, we propose a discrete mereotopology (DM) based approach to enable representation and querying of spatiotemporal information from a series of multitemporal RS images that are acquired during a flood disaster event. We represent this spatiotemporal information using a semantic model called Dynamic Flood Ontology (DFO). To establish the effectiveness and applicability of the proposed approach, spatiotemporal queries relevant during an urban flood scenario such as, show me road segments that were partially flooded during the time interval t1 have been demonstrated with promising results.

1 July 2018
Verified
Conference Paper
Closed access
APAbhishek Potnis
Abhishek Potnis
SDSurva S. Durbha
Surva S. Durbha
KKKuldeep R. Kurte
Kuldeep R. Kurte
Numerous remote sensing applications - flood monitoring, forest fires monitoring, earthquake analysis etc. require users to query satellite images based on their content. Such requirements have led to the evolution of Content-based Image Information Mining Systems over the last decade. Recent developments in the area of Image Information Mining(IIM) are geared towards bridging the gap between low level image features and higher-level semantics. This research focuses on improving the semantic understanding of a remote sensing scene during the flood disaster from a spatio-contextual standpoint. During a flood occurrence, it is crucial to understand the flood inundation and receding patterns in context to the spatial configurations of the land-use/land-cover in the flooded regions. This study focuses on bridging the spatio-contextual semantic gap in understanding of the remote sensing imagery during a flood, thereby attempting to improve the machine interpretability of a flood remote sensing imagery. In this regard, the Flood Scene Ontology (FSO) has been developed to mine the topological and directional knowledge in context to the flood disaster phenomenon. The FSO is envisaged to form the basis for developing applications that would utilize the spatio-contextual semantics of the flood disaster to aid in the Disaster Assessment and Management process. This paper describes the conceptual framework that was developed to address the same.