Search Results for author: Anand Rangarajan

Found 34 papers, 7 papers with code

Spatiotemporally adaptive compression for scientific dataset with feature preservation -- a case study on simulation data with extreme climate events analysis

no code implementations6 Jan 2024 Qian Gong, Chengzhu Zhang, Xin Liang, Viktor Reshniak, Jieyang Chen, Anand Rangarajan, Sanjay Ranka, Nicolas Vidal, Lipeng Wan, Paul Ullrich, Norbert Podhorszki, Robert Jacob, Scott Klasky

Additionally, we integrate spatiotemporal feature detection with data compression and demonstrate that performing adaptive error-bounded compression in higher dimensional space enables greater compression ratios, leveraging the error propagation theory of a transformation-based compressor.

Data Compression

A Spatiotemporal Correspondence Approach to Unsupervised LiDAR Segmentation with Traffic Applications

no code implementations23 Aug 2023 Xiao Li, Pan He, Aotian Wu, Sanjay Ranka, Anand Rangarajan

We address the problem of unsupervised semantic segmentation of outdoor LiDAR point clouds in diverse traffic scenarios.

Clustering Pseudo Label +3

An Efficient Semi-Automated Scheme for Infrastructure LiDAR Annotation

no code implementations25 Jan 2023 Aotian Wu, Pan He, Xiao Li, Ke Chen, Sanjay Ranka, Anand Rangarajan

Specifically, we introduce a human-in-the-loop schema in which annotators recursively fix and refine annotations imperfectly predicted by our tool and incrementally add them to the training dataset to obtain better SOT and MOT models.

Autonomous Driving Multi-Object Tracking +4

Scalable Hybrid Learning Techniques for Scientific Data Compression

1 code implementation21 Dec 2022 Tania Banerjee, Jong Choi, Jaemoon Lee, Qian Gong, Jieyang Chen, Scott Klasky, Anand Rangarajan, Sanjay Ranka

Data compression is becoming critical for storing scientific data because many scientific applications need to store large amounts of data and post process this data for scientific discovery.

Data Compression Video Compression

Expressing linear equality constraints in feedforward neural networks

1 code implementation8 Nov 2022 Anand Rangarajan, Pan He, Jaemoon Lee, Tania Banerjee, Sanjay Ranka

Elimination of the auxiliary variables leads to a dual minimization problem on the Lagrange multipliers introduced to satisfy the linear constraints.

Multi-Label Classification

Self-Supervised Robust Scene Flow Estimation via the Alignment of Probability Density Functions

no code implementations23 Mar 2022 Pan He, Patrick Emami, Sanjay Ranka, Anand Rangarajan

Scene flow estimation is therefore converted into the problem of recovering motion from the alignment of probability density functions, which we achieve using a closed-form expression of the classic Cauchy-Schwarz divergence.

Self-Supervised Learning Self-supervised Scene Flow Estimation

Learning Scene Dynamics from Point Cloud Sequences

no code implementations16 Nov 2021 Pan He, Patrick Emami, Sanjay Ranka, Anand Rangarajan

Our experimental evaluation confirms that recurrent processing of point cloud sequences results in significantly better SSFE compared to using only two frames.

Scene Flow Estimation Temporal Sequences

Generating Scenes with Latent Object Models

no code implementations29 Sep 2021 Patrick Emami, Pan He, Sanjay Ranka, Anand Rangarajan

We introduce a structured latent variable model that learns the underlying data-generating process for a dataset of scenes.

Object Re-Ranking +3

Domain-guided Machine Learning for Remotely Sensed In-Season Crop Growth Estimation

no code implementations24 Jun 2021 George Worrall, Anand Rangarajan, Jasmeet Judge

Results from this study exhibit both the viability of NNs in crop growth stage estimation (CGSE) and the benefits of using domain knowledge.

BIG-bench Machine Learning Time Series Analysis

Efficient Iterative Amortized Inference for Learning Symmetric and Disentangled Multi-Object Representations

1 code implementation7 Jun 2021 Patrick Emami, Pan He, Sanjay Ranka, Anand Rangarajan

Unsupervised multi-object representation learning depends on inductive biases to guide the discovery of object-centric representations that generalize.

Disentanglement Object

Neural data compression for physics plasma simulation

no code implementations ICLR Workshop Neural_Compression 2021 Jong Choi, Michael Churchill, Qian Gong, Seung-Hoe Ku, Jaemoon Lee, Anand Rangarajan, Sanjay Ranka, Dave Pugmire, CS Chang, Scott Klasky

We present a VAE-based data compression method, called VAe Physics Optimized Reduction (VAPOR), to compress scientific data while preserving physics constraints.

Data Compression

SparsePipe: Parallel Deep Learning for 3D Point Clouds

no code implementations27 Dec 2020 Keke Zhai, Pan He, Tania Banerjee, Anand Rangarajan, Sanjay Ranka

Besides, it suitably partitions the model when the GPUs are heterogeneous such that the computing is load-balanced with reduced communication overhead.

A Unified Framework for Multiclass and Multilabel Support Vector Machines

no code implementations25 Mar 2020 Hoda Shajari, Anand Rangarajan

We propose a straightforward extension to the SVM to cope with multiclass and multilabel classification problems within a unified framework.

Intelligent Intersection: Two-Stream Convolutional Networks for Real-time Near Accident Detection in Traffic Video

no code implementations4 Jan 2019 Xiaohui Huang, Pan He, Anand Rangarajan, Sanjay Ranka

In this paper, we propose a two-stream Convolutional Network architecture that performs real-time detection, tracking, and near accident detection of road users in traffic video data.

Multiple Object Tracking Object +2

Visual Explanations From Deep 3D Convolutional Neural Networks for Alzheimer's Disease Classification

1 code implementation7 Mar 2018 Chengliang Yang, Anand Rangarajan, Sanjay Ranka

Comparative analysis show that the sensitivity analysis based approach has difficulty handling loosely distributed cerebral cortex, and approaches based on visualization of activations are constrained by the resolution of the convolutional layer.

Classification General Classification +2

Global Model Interpretation via Recursive Partitioning

1 code implementation11 Feb 2018 Chengliang Yang, Anand Rangarajan, Sanjay Ranka

To generate the interpretation tree, a unified process recursively partitions the input variable space by maximizing the difference in the average contribution of the split variable between the divided spaces.

BIG-bench Machine Learning

A Gaussian mixture model representation of endmember variability in hyperspectral unmixing

1 code implementation29 Sep 2017 Yuan Zhou, Anand Rangarajan, Paul D. Gader

We show, given the GMM starting premise, that the distribution of the mixed pixel (under the linear mixing model) is also a GMM (and this is shown from two perspectives).

Hyperspectral Unmixing

A Category Space Approach to Supervised Dimensionality Reduction

no code implementations27 Oct 2016 Anthony O. Smith, Anand Rangarajan

To this end, we set up a model in which each class is represented as a 1D subspace of the vector space formed by the features.

Supervised dimensionality reduction

A Compositional Approach to Language Modeling

no code implementations1 Apr 2016 Kushal Arora, Anand Rangarajan

Traditional language models treat language as a finite state automaton on a probability space over words.

Language Modelling Sentence

A Grassmannian Graph Approach to Affine Invariant Feature Matching

no code implementations28 Jan 2016 Mark Moyou, John Corring, Adrian Peter, Anand Rangarajan

In this work, we present a novel and practical approach to address one of the longstanding problems in computer vision: 2D and 3D affine invariant feature matching.

Spatial Scaling of Satellite Soil Moisture using Temporal Correlations and Ensemble Learning

no code implementations21 Jan 2016 Subit Chakrabarti, Jasmeet Judge, Tara Bongiovanni, Anand Rangarajan, Sanjay Ranka

A novel algorithm is developed to downscale soil moisture (SM), obtained at satellite scales of 10-40 km by utilizing its temporal correlations to historical auxiliary data at finer scales.

Ensemble Learning regression

Disaggregation of SMAP L3 Brightness Temperatures to 9km using Kernel Machines

no code implementations20 Jan 2016 Subit Chakrabarti, Tara Bongiovanni, Jasmeet Judge, Anand Rangarajan, Sanjay Ranka

In this study, a machine learning algorithm is used for disaggregation of SMAP brightness temperatures (T$_{\textrm{B}}$) from 36km to 9km.

Image Segmentation Semantic Segmentation

Contrastive Entropy: A new evaluation metric for unnormalized language models

no code implementations3 Jan 2016 Kushal Arora, Anand Rangarajan

In this paper, we address the last problem and propose a new discriminative entropy based intrinsic metric that works for both traditional word level models and unnormalized language models like sentence level models.

Language Modelling Sentence

A spatial compositional model (SCM) for linear unmixing and endmember uncertainty estimation

no code implementations30 Sep 2015 Yuan Zhou, Anand Rangarajan, Paul Gader

In this paper, we show that NCM can be used for calculating the uncertainty of the estimated endmembers with spatial priors incorporated for better unmixing.

Hyperspectral Unmixing

Downscaling Microwave Brightness Temperatures Using Self Regularized Regressive Models

no code implementations30 Jan 2015 Subit Chakrabarti, Jasmeet Judge, Anand Rangarajan, Sanjay Ranka

A novel algorithm is proposed to downscale microwave brightness temperatures ($\mathrm{T_B}$), at scales of 10-40 km such as those from the Soil Moisture Active Passive mission to a resolution meaningful for hydrological and agricultural applications.

Clustering

Disaggregation of Remotely Sensed Soil Moisture in Heterogeneous Landscapes using Holistic Structure based Models

no code implementations30 Jan 2015 Subit Chakrabarti, Jasmeet Judge, Anand Rangarajan, Sanjay Ranka

The KLD of the disaggregated estimates generated by the SRRM is at least four orders of magnitude lower than those for the PRI disaggregated estimates, while the computational time needed was reduced by three times.

Clustering

A Riemannian Framework for Matching Point Clouds Represented by the Schrodinger Distance Transform

no code implementations CVPR 2014 Yan Deng, Anand Rangarajan, Stephan Eisenschenk, Baba C. Vemuri

In this paper, we use the well known Riemannian framework never before used for point cloud matching, and present a novel matching algorithm.

set matching

A fast eikonal equation solver using the Schrodinger wave equation

no code implementations8 Mar 2014 Karthik S. Gurumoorthy, Adrian M. Peter, Birmingham Hang Guan, Anand Rangarajan

In our framework, a solution to the eikonal equation is obtained in the limit as Planck's constant $\hbar$ (treated as a free parameter) tends to zero of the solution to the corresponding linear Schr\"odinger equation.

Gradient density estimation in arbitrary finite dimensions using the method of stationary phase

no code implementations13 Nov 2012 Karthik S. Gurumoorthy, Anand Rangarajan, John Corring

We prove that the density function of the gradient of a sufficiently smooth function $S : \Omega \subset \mathbb{R}^d \rightarrow \mathbb{R}$, obtained via a random variable transformation of a uniformly distributed random variable, is increasingly closely approximated by the normalized power spectrum of $\phi=\exp\left(\frac{iS}{\tau}\right)$ as the free parameter $\tau \rightarrow 0$.

Density Estimation

A new variational principle for the Euclidean distance function: Linear approach to the non-linear eikonal problem

no code implementations13 Dec 2011 Karthik S. Gurumoorthy, Anand Rangarajan

In other words, when $S$ and $\phi$ are related by $\phi = \exp \left(-\frac{S}{\tau} \right)$ and $\phi$ satisfies a specific linear differential equation corresponding to the extremum of a variational problem, we obtain the approximate Euclidean distance function $S = -\tau \log(\phi)$ which converges to the true solution in the limit as $\tau \rightarrow 0$.

Cannot find the paper you are looking for? You can Submit a new open access paper.