index.html

---
layout: default
tags: about
---
<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.4.1/jquery.min.js"></script>
<script type="text/javascript">
function readMore() {
    $('#readMore').hide();
    $('#more').show();
}
function readLess() {
    $('#readMore').show();
    $('#more').hide();
}
</script>

<img src="images/profile/adi_ny_crop.png" alt="Aditya Agarwal" width="240" style="float: right; padding: 20px; border-radius: 50%;" />

<div class="bio" style="text-align:justify">

<p>
  I am a first year EECS PhD student at MIT advised by professors <a href="https://people.csail.mit.edu/lpk/" class="uline-special">Leslie Kaelbling</a> and <a href="https://people.csail.mit.edu/tlp/" class="uline-special">Tomas Lozano-Perez</a>. I am part of the <a href="https://lis.csail.mit.edu/" class="uline-special">Learning and Intelligent Systems (LIS)</a> Group, part of the broader <a href="https://www.csail.mit.edu/" class="uline-special">CSAIL</a> <a href="https://ei.csail.mit.edu/" class="uline-special">Embodied Intelligence (EI)</a> Group. 
  I am broadly interested in problems at the intersection of robotic perception and TAMP. Previously, I was a visiting researcher at <a href="https://montrealrobotics.ca/" class="uline-special">REAL</a> at <a href="https://www.umontreal.ca/en/" class="uline-special">Université de Montréal</a> associated with <a href="https://mila.quebec/en/" class="uline-special">Mila</a>. I was advised by Prof. <a href="https://liampaull.ca/" class="uline-special">Liam Paull</a>.
  <!-- I am a research intern at the <a href="https://montrealrobotics.ca/" class="uline-special">Robotics and Embodied AI Lab (REAL)</a> at <a href="" class="uline-special">Université de Montréal</a> and <a href="https://mila.quebec/en/" class="uline-special">Mila</a>.  -->
  <!-- I work on representation learning for robotics systems, supervised by professors <a href="https://liampaull.ca/" class="uline-special">Liam Paull</a> (UdeM) and <a href="http://www.cs.toronto.edu/~florian/" class="uline-special">Florian Shkurti</a> (UofT).  -->
  <!-- I will be joining the <a href="https://lis.csail.mit.edu/" class="uline-special">Learning and Intelligent Systems (LIS) Group</a> as a PhD student in EECS at <a href="https://www.csail.mit.edu/" class="uline-special">MIT CSAIL</a> this fall.  -->
  <!-- I will work at the intersection of perception and task & motion planing for robotics systems supervised by professors <a href="https://people.csail.mit.edu/lpk/" class="uline-special">Leslie Pack Kaelbling</a> and <a href="https://people.csail.mit.edu/tlp/" class="uline-special">Tomas Lozano-Perez</a>, with the overarching goal of building general-purpose and autonomous robots that can seamlessly integrate with humans.  -->
</p>

<p>
  I completed my Masters by research from IIIT Hyderabad, and was advised by professors <a href="https://scholar.google.com/citations?user=U9dH-DoAAAAJ&hl=en" class="uline-special">C V Jawahar</a> and <a href="https://vinaypn.github.io/" class="uline-special">Vinay Namboodiri</a> in the computer vision (<a href="https://cvit.iiit.ac.in/" class="uline-special">CVIT</a>) lab and by professors <a href="https://scholar.google.co.in/citations?user=QDuPGHwAAAAJ&hl=en" class="uline-special">Madhava Krishna</a> and <a href="https://cs.brown.edu/people/ssrinath/" class="uline-special">Srinath Sridhar</a> (Brown University) in the robotics (<a href="https://robotics.iiit.ac.in/" class="uline-special">RRC</a>) lab.
  <!-- I completed my MS by research from <a href="https://www.iiit.ac.in/" class="uline-special">IIIT Hyderabad</a> supervised by professors <a href="https://scholar.google.com/citations?user=U9dH-DoAAAAJ&hl=en" class="uline-special">C V Jawahar</a> and <a href="https://vinaypn.github.io/" class="uline-special">Vinay Namboodiri</a> in the computer vision (<a href="https://cvit.iiit.ac.in/" class="uline-special">CVIT</a>) lab and by Prof. <a href="https://scholar.google.co.in/citations?user=QDuPGHwAAAAJ&hl=en" class="uline-special">Madhava Krishna</a> in the robotics (<a href="https://robotics.iiit.ac.in/" class="uline-special">RRC</a>) lab.  -->
  <!-- I am an MS by research student at IIIT Hyderabad. I am advised by <a href="https://scholar.google.com/citations?user=U9dH-DoAAAAJ&hl=en" class="uline-special">Prof. C V Jawahar</a> and <a href="https://vinaypn.github.io/" class="uline-special">Prof. Vinay Namboodiri</a> at the computer vision (<a href="https://cvit.iiit.ac.in/" class="uline-special">CVIT</a>) Lab, and by <a href="https://scholar.google.co.in/citations?user=QDuPGHwAAAAJ&hl=en" class="uline-special">Prof. Madhava Krishna</a> at the robotics (<a href="https://robotics.iiit.ac.in/" class="uline-special">RRC</a>) Lab of IIIT Hyderabad. -->
  My work spanned the broad areas of talking-face generation, video understanding and generation, and robotic perception & manipulation.
  <!-- My work spanned the areas of 3D shape completion, video understanding, implicit representations, robotic manipulation, and talking-face generation. -->
</p>
<!-- <p>
  My research interests lie broadly at the intersection of <strong>computer vision</strong> and <strong>robotics</strong>. 
  My long-term goal is to design task-driven representations of the 3D world that can enable embodied agents to perceive and interact with the 3D environment intelligently, and perform various robotic tasks efficiently. 
</p> -->
<br>
<p>
  Apart from academia, I spent time as a <a href="https://www.youtube.com/watch?v=-ZEXU20tkFw" class="uline-special">Software Engineer</a> at Microsoft India, and worked for Bing (People Also Ask feature) and Azure (<a href="" class="uline-special">Education</a> and Healthcare initiatives) organizations. My work was covered in the press. 
  <!-- Previously, I was a <a href="https://www.youtube.com/watch?v=-ZEXU20tkFw" class="uline-special">Software Engineer at Microsoft</a> India in the People Also Ask (PAA) team. I worked on techniques in deep learning and NLP to show a block of related questions and answers for a user query on Bing's search page.  -->
</p>
<br>
<p>
  I completed my Bachelors from <a href="https://pes.edu/" class="uline-special">PES University</a> (formerly PESIT) Bangalore in Computer Science. I worked in the areas of sound event detection and localization, and spent a considerable amount of time interning at 
  The <a href="https://www.ucalgary.ca/" class="uline-special">University of Calgary</a> (through <a href="https://www.mitacs.ca/en/programs/globalink/globalink-research-internship" class="uline-special">MITACS</a>) with <a href="https://contacts.ucalgary.ca/info/enel/profiles/168-42833" class="uline-special">Prof. Mike Smith</a> and <a href="https://www.microsoft.com/en-us/research/lab/microsoft-research-india/" class="uline-special">Microsoft Research India</a>.
  <!-- I completed my Bachelors from <a href="https://pes.edu/" class="uline-special">PES University</a> (formerly PESIT) Bangalore in Computer Science. At PESIT, I worked in the areas of sound event detection and localization.  -->
  <!-- I also spent a summer as a MITACS research intern at the <a href="https://www.ucalgary.ca/" class="uline-special">University of Calgary</a> on localizing an audio noise nuisance called the Ranchlands Hum, supervised by Prof. <a href="http://people.ucalgary.ca/~smithmr/MRSmith_DepartmentalWeb/indexMain.htm" class="uline-special">Mike Smith</a>,  -->
  <!-- and a year as an intern at <a href="https://www.microsoft.com/en-us/research/lab/microsoft-research-india/" class="uline-special">Microsoft Research India</a>, working in the areas of blended learning and AI in healthcare.  -->
  <!-- I spent a wonderful summer in Canada as a MITACS intern at the <a href="https://www.ucalgary.ca/" class="uline-special">University of Calgary</a> on characterizing and localizing an audio noise nuisance, and was advised by <a href="http://people.ucalgary.ca/~smithmr/MRSmith_DepartmentalWeb/indexMain.htm" class="uline-special">Prof. Mike Smith</a>.  -->
  <!-- on characterizing and localizing an audio nuisance dubbed the Ranchlands Hum. -->
  <!-- I also spent a year as a research intern at <a href="https://www.microsoft.com/en-us/research/lab/microsoft-research-india/" class="uline-special">Microsoft Research Labs</a>, working in the areas of blended learning and AI in healthcare. -->
</p>

<p><strong>Research Interests: </strong><span style="color:gray">My research interests lie broadly at the intersection of computer vision and robotics. My goal is to integrate representation learning with task & motion planning to achieving general-purpose robot autonomy.</span></p>
<!-- <p><strong>Interest in Robotics: </strong><span style="color:gray">I focus on improving the perception, grasping, and navigation,  capabilities of robotics systems, aiming to develop agents that can intelligently perceive and reason about the environment and perform tasks efficiently. Specifically, I work in <a href="https://bipashasen.github.io/scarp/" class="uline-special"><span style="color:red">3D scene completion</span></a>, implicit neural representations for 3D shapes, and exploring synergies between different motion primitives for enabling complex robotics behavior. I am also interested in grounding natural language instructions with the robot's underlying representation of the 3D world to allow behaviors such as language-guided <a href="https://arxiv.org/pdf/2205.04090.pdf" class="uline-special"><span style="color:red">tabletop manipulation</span></a>, voice-based semantic navigation, and human-robotic interaction.</span></p> -->
  <!-- In robotics, I focus on improving the perception, grasping, and navigation capabilities of robotics systems with the eventual aim of developing agents that can perceive and reason about the environment intelligently and perform tasks efficiently. Specifically, I work in the areas of 3D scene completion, implicit neural representations for 3D shapes, and exploring synergies between different motion primitives for enabling complex robotics behavior. 
  I am also interested in grounding natural language instructions with the robot's underlying representation of the 3D world to enable behaviors such as language-guided tabletop manipulation, voice-based semantic navigation, and human-robotic interaction. -->
<!-- <p><strong>Interest in Computer Vision: </strong><span style="color:gray"> My research in computer vision has focused on generative modeling, <a href="/INRV/" class="uline-special"><span style="color:red">video representation and understanding</span></a>, and <a href="http://cvit.iiit.ac.in/research/projects/cvit-projects/faceoff" class="uline-special"><span style="color:red">face-swapping</span></a>. My recent work was toward designing an alternate representation space for videos where videos are encoded as implicit functions. In face-swapping, I introduced a new line of research towards video face-swapping that tackles pressing challenges in the moviemaking industry. I am also profoundly concerned about the potential of deepfakes misuse. My recent work is toward deepfake detection and attribution with the eventual aim of releasing an ethical framework for training and releasing generative models in the wild.</span></p> -->
  <!-- In CV, my research has been focused towards generative modeling, video representation and understanding, and face-swapping. My recent work was toward designing an alternate representation space for videos where videos are encoded as implicit functions.  -->
  <!-- Conditioning videos as INRs offers several possiblities such as video inpainting, superresolution, completion, denoising without explicitly enforcing these capabilities.  -->
  <!-- In face-swapping, I introduced a new line of research towards video face-swapping, that tackles pressing challenges in the moviemaking industry. I am also deeply concerned about the potential of deepfakes misue. My very recent work is toward deepfake detection and attribution with the eventual aim of releasing an ethical framework for training and releasing generative models in the wild.  -->

<!-- <p><strong>Research Interests: <span style="color:gray">My interests lie in designing improved representations of the 3D and 2D worlds that are most suitable for enabling diverse tasks and applications. I envision a scenario where robots perceive and interact with the environment seamlessly to perform tasks such as navigation and manipulation. 
  In robotics, I am working on 3D scene completion and denoising NeRFs to improve robotics perception and grasping; and on exploring synergies between different motion primitives toward robotic planning and manipulation. 
  I am also interested in grounding natural language instructions with the robot's underlying representation of the 3D world to enable behaviors such as voice-based semantic navigation, language-guided robotic tabletop manipulation, and human-robotic interaction.
  On the vision side, I am interested in generative modeling, video understanding, and face manipulation. Existing video generation works rely on learning temporally coherent trajectories in the learned space of image-based generators which is highly restrictive. 
  I am working on learning an efficient representation space for videos that can enable several video-based generative tasks to augment the experiences offered by content streaming platforms. 
  I am also working on an ethical framework for releasing generative models. </span></strong></p> -->

<!-- <p><strong>Contact: </strong>I am always happy to discuss and collaborate on research ideas. Email me at <a href="mailto:skymanaditya1@gmail.com" class="uline">skymanaditya1@gmail.com</a></p> -->

<br/>

<div class="container">
  <div class="row" style="text-align:center;">
        <div class="col">
          <a href="https://pes.edu/"><img src="images/affiliations/resized/pes.jpg" style="max-height:60px;width:60%"></a>
        </div>
        <div class="col" style="text-align:center;">
          <a href="https://www.iiit.ac.in/"><img src="images/affiliations/resized/iiith.png" style="max-height:60px;width:60%"></a>
        </div>
        <div class="col" style="text-align:center;">
          <a href="https://www.mit.edu/"><img src="images/affiliations/mit_square.png" style="max-height:60px;width:60%"></a>
        </div>
        <div class="col" style="text-align:center;">
          <a href="https://www.microsoft.com/en-us/research/lab/microsoft-research-india/"><img src="images/affiliations/resized/microsoft.png" style="width:60%;max-height:60px"></a>
        </div>
        <div class="col">
          <a href="https://www.ucalgary.ca/"><img src="images/affiliations/resized/uofc.png" style="width:60%;max-height:60px"></a>
        </div>
        <div class="col">
          <a href="https://www.vmware.com/in.html/"><img src="images/affiliations/vmware.jpeg" style="width:60%;max-height:60px"></a>
        </div>
        <div class="col" style="text-align:center;">
          <a href="https://www.microsoft.com/en-in"><img src="images/affiliations/resized/microsoft.png" style="width:60%;max-height:60px"></a>
        </div>
        <div class="col" style="text-align:center;">
          <a href="https://www.umontreal.ca/en/"><img src="images/affiliations/umontreal_logo.jpeg" style="width:60%;max-height:60px"></a>
        </div>
  </div>
  <div class="row" style="text-align:center;">
        <div class="col">
          <div style="padding:10px"><h6>2013 - 2017</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>2021 - 2023</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>2023 - Current</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>Spring <br>'15 & '16</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>Summer '16</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>Fall '17</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>2018 - 2021</h6></div>
        </div>
        <div class="col">
          <div style="padding:10px"><h6>Summer '23</h6></div>
        </div>
  </div>
</div>

<hr/>
<br/>

<div id="research">
<h2>
  <a name="research">Publications</a>
</h2>
<br/>

<table width="100%" align="center" valign="middle" cellspacing="0" cellpadding="0" style="border-collapse: collapse;">
    <tbody>
      <tr>
        <td width="30%">
          <div class="one" style="text-align:center;">
              <img src="images/projects/concept_graphs_arxiv2023/banner.png" style="max-height: 400px;">
          </div>
        </td>
        <td valign="top" width="70%">
          <h5>
            ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning
          </h5>
          <p class="authors" style="display:inline">Qiao Gu*, Ali Kuwajerwala*, Sacha Morin*, Krishna Murthy Jatavallabhula*, Bipasha Sen, <p class="self" style="display:inline">Aditya Agarwal</p>, <p class="authors" style="display:inline">Kirsty Ellis, Celso Miguel de Melo, Corban Rivera, William Paul, Rama Chellapa, Chuang Gan, Joshua B. Tenenbaum, Antonio Torralba, Florian Shkurti, Liam Paull</p>
        </p>
           <p>
            <!-- <i><span style="color:gray"></span>Under Review</i>, <a href="https://www.icra2023.org/" class="uline-special"><span style="color:red">ICRA 2023</span></a> -->
            <a href="images/projects/concept_graphs_arxiv2023/2024_icra_concept_graphs.pdf" class="uline-special"><span style="color:red">Under Review 2023</span></a>
          </p>
          <p>
            <!-- <a href="https://arxiv.org/pdf/2301.07213.pdf">Paper / </a> -->
            <a href="images/projects/concept_graphs_arxiv2023/2024_icra_concept_graphs.pdf">Paper / </a>
            <a href="https://concept-graphs.github.io/">Project Page / </a>
            <!-- <a href="https://bipashasen.github.io/scarp/">Project Page / </a> -->
            <!-- <a href="https://www.youtube.com/watch?v=o2PuRVZ3jJA">Short Video / </a> -->
            <a href="">Code (Coming Soon) / </a>
            <a href="">Video (Coming Soon)</a>
            <!-- <a href="images/projects/scarp_icra2023/poster.pdf">Poster / </a> -->
            <!-- <a href="https://youtu.be/lyEc991wyTI">Long Video</a> -->
          </p>
          <!-- &nbsp;<p style="color:red">More details coming soon!</p> -->
          <p>
            We propose ConceptGraphs, an open-vocabulary graph-structured representations for 3D scenes, built by leveraging 2D foundation
            models and fusing their output to 3D by multiview association. The resulting representations generalize to novel semantic classes, 
            without the need to collect large 3D datasets or finetune models. The utility of this representation is demonstrated through downstream robotic planning tasks.
          </p>
        </td>
      </tr>
      <tr>
      <tr>
        <td width="30%">
          <div class="one" style="text-align:center;">
              <img src="images/projects/edmp_arxiv2023/banner.png" style="max-height: 400px;">
          </div>
        </td>
        <td valign="top" width="70%">
          <h5>
            EDMP: Ensemble-of-costs-guided Diffusion for Motion Planning
          </h5>
          <p class="authors" style="display:inline">Kallol Saha*, Vishal Mandadi*, Jayaram Reddy*, Ajit Srikanth, <p class="self" style="display:inline">Aditya Agarwal</p>, <p class="authors" style="display:inline">Bipasha Sen, Arun Singh, Madhava Krishna</p>
        </p>
           <p>
            <!-- <i><span style="color:gray"></span>Under Review</i>, <a href="https://www.icra2023.org/" class="uline-special"><span style="color:red">ICRA 2023</span></a> -->
            <a href="https://arxiv.org/pdf/2309.11414.pdf" class="uline-special"><span style="color:red">Under Review 2023</span></a>
          </p>
          <p>
            <!-- <a href="https://arxiv.org/pdf/2301.07213.pdf">Paper / </a> -->
            <a href="https://arxiv.org/pdf/2309.11414.pdf">Paper / </a>
            <a href="https://ensemble-of-costs-diffusion.github.io/">Project Page / </a>
            <!-- <a href="https://bipashasen.github.io/scarp/">Project Page / </a> -->
            <!-- <a href="https://www.youtube.com/watch?v=o2PuRVZ3jJA">Short Video / </a> -->
            <a href="">Code (Coming Soon) / </a>
            <a href="https://youtu.be/F2UI0UNsdjo">Video</a>
            <!-- <a href="images/projects/scarp_icra2023/poster.pdf">Poster / </a> -->
            <!-- <a href="https://youtu.be/lyEc991wyTI">Long Video</a> -->
          </p>
          <!-- &nbsp;<p style="color:red">More details coming soon!</p> -->
          <p>
            We propose EDMP for motion planning, that combines the strengths of classical motion planning (offering remarkable adaptability) and deep-learning-based motion planning (prior understanding over diverse valid trajectories). 
            Our diffusion-based network is trained on a set of diverse kinematically valid trajectories. For any new scene at the time of inference,
            we compute scene-specific costs such as "collision cost" and guide the generation of valid trajectories that satisfy scene-specific constraints.
          </p>
        </td>
      </tr>
      <tr>
        <td width="30%">
          <div class="one" style="text-align:center;">
              <img src="images/projects/hypnerf_arxiv2023/project_picture.png" style="max-height: 400px;">
          </div>
        </td>
        <td valign="top" width="70%">
          <h5>
            HyP-NeRF: Learning Improved NeRF Priors using a HyperNetwork
          </h5>
          <p class="authors" style="display:inline">Bipasha Sen*, Gaurav Singh*, </p><p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Madhava Krishna, Srinath Sridhar</p>
        </p>
           <p>
            <!-- <i><span style="color:gray"></span>Under Review</i>, <a href="https://www.icra2023.org/" class="uline-special"><span style="color:red">ICRA 2023</span></a> -->
            <a href="https://arxiv.org/pdf/2306.06093.pdf" class="uline-special"><span style="color:red">NeurIPS 2023</span></a>
          </p>
          <p>
            <!-- <a href="https://arxiv.org/pdf/2301.07213.pdf">Paper / </a> -->
            <a href="images/projects/hypnerf_arxiv2023/HyP-NeRF.pdf">Paper / </a>
            <!-- <a href="https://bipashasen.github.io/scarp/">Project Page / </a> -->
            <!-- <a href="https://www.youtube.com/watch?v=o2PuRVZ3jJA">Short Video / </a> -->
            <a href="https://github.com/skymanaditya1/HyP-NeRF">Code (Coming Soon) / </a>
            <a href="https://www.youtube.com/watch?v=40JOIWJAvGs">Video</a>
            <!-- <a href="images/projects/scarp_icra2023/poster.pdf">Poster / </a> -->
            <!-- <a href="https://youtu.be/lyEc991wyTI">Long Video</a> -->
          </p>
          <!-- &nbsp;<p style="color:red">More details coming soon!</p> -->
          <p>
            We propose HyP-NeRF, a latent conditioning method for learning generalizable category-level NeRF priors using hypernetworks. 
            We use hypernetworks to estimate both the weights and the multi-resolution hash encodings resulting in significant quality gains. 
            To further improve quality, we incorporate a denoise and finetune strategy that denoises images rendered from NeRFs estimated by the hypernetwork and finetunes it while retaining multiview consistency.
          </p>
        </td>
      </tr>
      <tr>
        <td width="30%">
          <div class="one" style="text-align:center;">
              <img src="images/projects/synergistic_case2023/teaser.png" style="max-height: 400px;">
          </div>
        </td>
        <td valign="top" width="70%">
          <h5>
            Disentangling Planning and Control for Non-prehensile Tabletop Manipulation
          </h5>
          <p class="authors" style="display:inline">Vishal Reddy Mandadi, ..., </p><p class="self" style="display:inline">Aditya Agarwal</p><p class="authors" style="display:inline">, ..., Madhava Krishna</p>
        </p>
           <p>
            <!-- <i><span style="color:gray"></span>Under Review</i>, <a href="https://www.icra2023.org/" class="uline-special"><span style="color:red">ICRA 2023</span></a> -->
            <a href="https://case2023.org/" class="uline-special"><span style="color:red">CASE 2023</span></a>
          </p>
          <p>
            <!-- <a href="https://arxiv.org/pdf/2301.07213.pdf">Paper / </a> -->
            <a href="">Paper (Coming Soon)/ </a>
            <!-- <a href="https://bipashasen.github.io/scarp/">Project Page / </a> -->
            <a href="">Video (Coming Soon)</a>
            <!-- <a href="https://github.com/vanhalen42/SCARP">Code / </a> -->
            <!-- <a href="images/projects/scarp_icra2023/poster.pdf">Poster / </a> -->
            <!-- <a href="https://youtu.be/lyEc991wyTI">Long Video</a> -->
            <!-- &nbsp;<p style="color:red">More details coming soon!</p> -->
          </p>
          <p>
            We propose a framework that disentangles planning and control for tabletop manipulation in unknown scenes 
            using a pushing-by-striking method (without tactile feedback) by explicitly modeling the object dynamics. 
            Our method consists of two components: an A* planner for path-planning and a low-level RL controller that models object dynamics. 
          </p>
        </td>
      </tr>
      <tr>
      <td width="30%">
        <div class="one" style="text-align:center;">
            <img src="images/projects/scarp_icra2023/banner.png" style="max-height: 400px;">
        </div>
      </td>
      <td valign="top" width="70%">
        <h5>
          SCARP: 3D Shape Completion in ARbitrary Poses for Improved Grasping
        </h5>
        <p class="authors" style="display:inline">Bipasha Sen*, </p><p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Gaurav Singh*, Brojeshwar B., Srinath Sridhar, Madhava Krishna</p>
      </p>
         <p>
          <!-- <i><span style="color:gray"></span>Under Review</i>, <a href="https://www.icra2023.org/" class="uline-special"><span style="color:red">ICRA 2023</span></a> -->
          <a href="https://www.icra2023.org/" class="uline-special"><span style="color:red">ICRA 2023</span></a>
        </p>
        <p>
          <!-- <a href="https://arxiv.org/pdf/2301.07213.pdf">Paper / </a> -->
          <a href="images/projects/scarp_icra2023/ICRA_2429_published.pdf">Paper / </a>
          <a href="https://bipashasen.github.io/scarp/">Project Page / </a>
          <a href="https://www.youtube.com/watch?v=o2PuRVZ3jJA">Short Video / </a>
          <a href="https://github.com/vanhalen42/SCARP">Code / </a>
          <a href="images/projects/scarp_icra2023/poster.pdf">Poster / </a>
          <a href="https://youtu.be/lyEc991wyTI">Long Video</a>
        </p>
        <p>
          We propose a mechanism for completing partial 3D shapes in arbitrary poses by learning a disentangled feature representation of pose and shape. 
          We rely on learning rotationally equivariant pose features and geometric shape features by training a multi-tasking objective. 
          SCARP improves the shape completion performance by 45% and grasp proposals by 71.2% over existing baselines. 
        </p>
      </td>
    </tr>
    <tr>
    <td width="30%">
      <div class="one" style="text-align:center;">
          <img src="images/projects/faceoff_wacv2023/teaser_images/faceoff1.png" style="max-height: 200px;">
      </div>
    </td>
    <td valign="top" width="70%">
      <h5>
        FaceOff: A Video-to-Video Face Swapping System
      </h5>
      <p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Bipasha Sen*, Rudrabha Mukhopadhyay, Vinay P. Namboodiri, C.V. Jawahar</p>
    </p>
      <p>
        <a href="https://wacv2023.thecvf.com/home" class="uline-special"><span style="color:red">WACV 2023</span></a>
        <!-- IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2023 -->
      </p>
      <p>
        <a href="https://openaccess.thecvf.com/content/WACV2023/papers/Agarwal_FaceOff_A_Video-to-Video_Face_Swapping_System_WACV_2023_paper.pdf">Paper / </a>
        <a href="http://cvit.iiit.ac.in/research/projects/cvit-projects/faceoff">Project Page / </a>
        <a href="https://www.youtube.com/watch?v=aNhs-mqMOcE">Video / </a>
        <a href="images/projects/faceoff_wacv2023/615-wacv-post.pdf">Poster / </a>
        <a href="https://github.com/skymanaditya1/FaceOff">Code / </a>
        <a href="images/projects/faceoff_wacv2023/0615-supp.pdf">Supplementary</a>
      </p>
      <p>
        We propose a novel direction of video-to-video (V2V) face-swapping that tackles a pressing challenge in the moviemaking industry: 
        swapping the actor's face and expressions on the face of their body double. 
        Existing face-swapping methods preserve only the identity of the source face without swapping the expressions. 
        In FaceOff, we swap the source's facial expressions along with the identity on the target's background and pose. 
      </p>
    </td>
  </tr>
  <tr>
  <td width="30%">
    <div class="one" style="text-align:center;">
        <img src="images/projects/moocs-lrt_wacv2023/teaser_images/lipreading2.png" style="max-height: 200px;">
    </div>
  </td>
  <td valign="top" width="70%">
    <h5>
      Towards MOOCs for Lipreading: Using Synthetic Talking Heads to Train Humans in Lipreading at Scale
    </h5>
    <!-- <p class="authors">
      <b>Aditya Agarwal*</b>, Bipasha Sen*, Rudrabha Mukhopadhyay, Vinay P. Namboodiri, C.V. Jawahar
    </p> -->
    <p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Bipasha Sen*, Rudrabha Mukhopadhyay, Vinay P. Namboodiri, C.V. Jawahar</p>
  </p>
    <p>
      <a href="https://wacv2023.thecvf.com/home" class="uline-special"><span style="color:red">WACV 2023</span></a>
      <!-- IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2023 -->
    </p>
    <p>
      <a href="https://openaccess.thecvf.com/content/WACV2023/papers/Agarwal_Towards_MOOCs_for_Lipreading_Using_Synthetic_Talking_Heads_To_Train_WACV_2023_paper.pdf">Paper / </a>
      <a href="http://cvit.iiit.ac.in/research/projects/cvit-projects/mooc-lip">Project Page / </a>
      <a href="https://youtu.be/tc2Pt6dyjjo">Video / </a>
      <a href="images/projects/moocs-lrt_wacv2023/720-wacv-post.pdf">Poster / </a>
      <!-- <a href="https://github.com/skymanaditya1/FaceOff">Code / </a> -->
      <a href="images/projects/moocs-lrt_wacv2023/0720-supp.pdf">Supplementary</a>
    </p>
    <p>
      Hard-of-hearing people rely on lipreading the mouth movements of the speaker to understand the spoken content. 
      In this work, we developed computer vision techniques and built upon existing AI models, such as TTS and talking-face generation, 
      to generate synthetic lipreading training content in any language.
    </p>
  </td>
  </tr>
  <tr>
  <td width="30%">
    <div class="one" style="text-align:center;">
        <img src="images/projects/inr-v_tmlr2022/teaser.png" style="max-height: 200px;">
    </div>
  </td>
  <td valign="top" width="70%">
    <h5>
      INR-V: A Continuous Representation Space for Video-based Generative Tasks
    </h5>
    <p class="authors" style="display:inline">Bipasha Sen*, </p><p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Vinay P. Namboodiri, C.V. Jawahar</p>
    </p>
    <p>
      <a href="https://www.jmlr.org/tmlr/" class="uline-special"><span style="color:red">TMLR 2022</span></a>
      <!-- Transactions on Machine Learning Research, TMLR 2022 -->
    </p>
    <p>
      <a href="https://openreview.net/pdf?id=aIoEkwc2oB">Paper / </a>
      <a href="https://openreview.net/forum?id=aIoEkwc2oB">OpenReview / </a>
      <!-- <a href="http://cvit.iiit.ac.in/research/projects/cvit-projects/inr-v">Project Page / </a> -->
      <a href="/INRV/">Project Page / </a>
      <a href="https://www.youtube.com/watch?v=ViIwnu5vcck">Video / </a>
      <a href="https://github.com/bipashasen/INRV">Code</a>
    </p>
    <p>
      Inspired by the recent works on parameterizing 3D shapes and scenes as Implicit Neural Representations (INRs), 
      we encode videos as INRs. We train a hypernetwork to learn a prior over these INR functions and propose two techniques, 
      i) Progressive Training and ii) Video-CLIP Regularization to stabilize hypernetwork training. 
      INR-V shows remarkable performance on several video-generative tasks on many benchmark datasets.
    </p>
    <!-- <p>
      *Denotes equal contribution
    </p> -->
  </td>
  </tr>
  <tr>
  <td width="30%">
    <div class="one" style="text-align:center;">
        <img src="images/projects/ocrtoc_icra2022/tabletop1.png" style="max-height: 200px;">
    </div>
  </td>
  <td valign="top" width="70%">
    <h5>
      Approaches and Challenges in Robotic Perception for Table-top Rearrangement and Planning
    </h5>
    <p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Bipasha Sen*, Shankara Narayanan V*, Vishal Reddy Mandadi*, Brojeshwar Bhowmick, K Madhava Krishna</p>
    </p>
    <p>
      <a href="https://rpal.cse.usf.edu/rgmc_icra2022/" class="uline-special"><span style="color:red">3<sup>rd</sup> in ICRA 2022 Open Cloud Table Organization Challenge</span></a>
      <!-- 3<sup>rd</sup> in ICRA 2022 Open Cloud Table Organization Challenge -->
    </p>
    <p>
      <a href="https://arxiv.org/pdf/2205.04090.pdf">Paper / </a>
      <a href="https://rpal.cse.usf.edu/rgmc_icra2022/">Competition / </a>
      <a href="https://youtu.be/5VSP49OB0ZI">Video / </a>
      <a href="images/projects/ocrtoc_icra2022/WeeklyPresentations.pdf">Slides / </a>
      <a href="https://github.com/skymanaditya1/ocrtoc_iiith/tree/jan30_sub_6dof">Code / </a>
      <a href="https://blogs.iiit.ac.in/rrc/">News1 / </a>
      <a href="https://www.iiit.ac.in/files/media/Sakshi-RRC.jpeg">News2</a>
    </p>
    <p>
      In this challenge, we proposed an end-to-end pipeline in ROS incorporating the perception and planning stacks to manipulate objects from their 
      initial configuration to a desired target configuration on a tabletop scene using a two-finger manipulator. 
      The pipeline involves the following steps - (1) 3D scene registration, (2) Object pose estimation, (3) Grasp generation, 
      (4) Task Planning, and (5) Motion Planning. 
    </p>
  </td>
  </tr>
  <tr>
    <td width="30%">
      <div class="one" style="text-align:center;">
          <img src="images/projects/als_bmvc2021/banner_images/image_overlay_white.png" style="max-height: 400px;">
      </div>
    </td>
    <td valign="top" width="70%">
      <h5>
        Personalized One-Shot Lipreading for an ALS Patient
      </h5>
      <p class="authors" style="display:inline">Bipasha Sen*, </p><p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Rudrabha Mukhopadhyay, Vinay P. Namboodiri, C.V. Jawahar</p>
      </p>
      <p>
        <a href="https://www.bmvc2021-virtualconference.com/" class="uline-special"><span style="color:red">BMVC 2021</span></a>
        <!-- British Machine Vision Conference, BMVC 2021 -->
      </p>
      <p>
        <a href="https://www.bmvc2021-virtualconference.com/assets/papers/1468.pdf">Paper / </a>
        <a href="https://www.youtube.com/watch?v=_famGVaem-8">Video</a>
      </p>
      <p>
        We tackled the challenge of lipreading medical patients in a one-shot setting. 
        There were two primary issues in training existing lipreading models - i) lipreading datasets had people suffering from no disabilities, 
        ii) lipreading datasets lacked medical words. We devised a variational encoder-based domain adaptation technique to adapt models 
        trained on large amounts of synthetic data to enable lipreading with one-shot real examples.
      </p>
    </td>
  </tr>
  <tr>
    <td width="30%">
      <div class="one" style="text-align:center;">
          <img src="images/projects/reed_slt2021/architecture.png" style="max-height: 400px;">
      </div>
    </td>
    <td valign="top" width="70%">
      <h5>
        REED: An Approach Towards Quickly Bootstrapping Multilingual Acoustic Models
      </h5>
      <p class="authors" style="display:inline">Bipasha Sen*, </p><p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Mirishkar Sai Ganesh, Anil Kumar Vuppala</p>
      </p>
      <p>
        <a href="http://2021.ieeeslt.org/" class="uline-special"><span style="color:red">SLT 2021</span></a>
        <!-- IEEE Spoken Language Technology Workshop, SLT 2021 -->
      </p>
      <p>
        <a href="images/projects/reed_slt2021/reed_paper.pdf">Paper / </a>
        <a href="images/projects/reed_slt2021/reed_presentation.pdf">Slides / </a>
        <a href="images/projects/reed_slt2021/MASR_Synapse.pdf">MLADS Paper</a>
      </p>
      <p>
        We tackled the problem of building a multilingual acoustic model in a low-resource setting. 
        We proposed a mechanism to bootstrap and validate the compatibility of multiple languages using CNNs operating directly on raw speech signals. 
        Our method improves training and inference times by 4X and 7.4X, respectively, with comparable WERs against RNN-based baseline systems. 
      </p>
    </td>
  </tr>
  <tr>
    <td width="30%">
      <div class="one" style="text-align:center;">
          <img src="images/projects/actionrecognition_isvc2020/architecture.png" style="max-height: 400px;">
      </div>
    </td>
    <td valign="top" width="70%">
      <h5>
        An Approach Towards Action Recognition using Part Based Hierarchical Fusion
      </h5>
      <p class="self" style="display:inline">Aditya Agarwal*</p><p class="authors" style="display:inline">, Bipasha Sen*</p>
      </p>
      <p>
        <a href="https://www.isvc.net/" class="uline-special"><span style="color:red">ISVC 2020</span></a>
        <!-- International Symposium on Visual Computing, ISVC 2020 -->
      </p>
      <p>
        <a href="images/projects/actionrecognition_isvc2020/humanactionrecognition_partbasedhierarchicalfusion.pdf">Paper / </a>
        <a href="images/projects/actionrecognition_isvc2020/slides_190.pdf">Slides / </a>
        <a href="images/projects/actionrecognition_isvc2020/PBAR_Synapse.pdf">MLADS Paper</a>
      </p>
      <p>
        The human body can be represented as an articulation of rigid and hinged joints, which can be combined to form the parts of the body. 
        In this work, we think of human actions as a collective action of these parts. 
        We propose a Hierarchical BiLSTM network to model the spatio-temporal dependencies of the motion by fusing the pose-based joint trajectories 
        in a part-based hierarchical fashion.
      </p>
    </td>
  </tr>
  <tr>
    <td width="30%">
      <div class="one" style="text-align:center;">
          <img src="images/projects/icacci2016/architecture.png" style="max-height: 400px;">
      </div>
    </td>
    <td valign="top" width="70%">
      <h5>
        Minimally Supervised Sound Event Detection using a Neural Network
      </h5>
      <p class="self" style="display:inline">Aditya Agarwal</p><p class="authors" style="display:inline">, Syed Munawwar Quadri, Savitha Murthy, Dinkar Sitaram</p>
      </p>
      <p>
        <a href="https://ieeexplore.ieee.org/xpl/conhome/7592392/proceeding" class="uline-special"><span style="color:red">ICACCI 2016</span></a>
        <!-- International Conference on Advances in Computing, Communications and Informatics, ICACCI 2016 -->
      </p>
      <p>
        <a href="images/projects/icacci2016/icacci2016_paper.pdf">Paper / </a>
        <a href="images/projects/icacci2016/icacci_poster2016.pdf">Poster / </a>
        <a href="https://github.com/skymanaditya1/Minimally-Supervised-Sound-Event-Detection">Code</a>
      </p>
      <p>
        We solve the task of polyphonic sound event detection by training on a minimally annotated dataset of single sounds. 
        Single sounds represented as MFCC features are used to train a neural network. 
        Polyphonic sounds are preprocessed using PCA and NMF, and source-separated sounds are inferred using the learned network. 
        Our system achieves reasonable accuracy of source separation and detection with minimal data.
      </p>
    </td>
  </tr>
  </tbody>
</table>
</div>

<br/>
<hr/>

<!-- <div id="talks">
<h2>News & Announcements</h2>
<table width="100%" align="center" valign="middle" cellspacing="0" cellpadding="0" style="border-collapse: collapse;">
    <tbody>

      <tr>
      <td width="20%">
        <h5>
          January, 2023
        </h5>
      </td>
      <td valign="middle" width="80%">
        <h5>
          I will be attending the Google Research Week 2023 in Bangalore from 29<sup>th</sup> January to 31<sup>st</sup> January.
        </h5>
      </td>
      </tr>

      <tr>
      <td width="20%">
        <h5>
          December, 2022
        </h5>
      </td>
      <td valign="middle" width="80%">
        <h5>
          I will be serving as a Reviewer at the Neural Fields Workshop at ICLR 2023 (NF2023).
        </h5>
      </td>
      </tr>

      <tr>
      <td width="20%">
        <h5>
          November, 2022
        </h5>
      </td>
      <td valign="middle" width="80%">
        <h5>
          I served as a Reviewer at ICRA 2023.
        </h5>
      </td>
      </tr>

      <tr>
      <td width="20%">
        <h5>
          September, 2020
        </h5>
      </td>
      <td valign="middle" width="80%">
        <h5>
        Gave a tutorial on Geometric Deep Learning and Graph Convolutional Networks (GCN)
        </h5>
        <p class="authors">
        Reading Group: Vision for Mobility and Safety
        <br>
        <a href="reports/GCN-Tutorial-Wed-meeting.pdf">Slides</a>
        </p>
      </td>
      </tr>

  </tbody>
</table>
</div> -->

<div id="news">
  <h2>News & Announcements</h2>
  <br />
  <ul>
  <!-- <li>
    <p>
      [Nov '23] Serving as reviewer for ICRA 2024 and CVPR 2024.
    </p>
  -->
  <li>
    <p>
      [Nov '23] Awarded <a href="https://nips.cc/" class="uline">NeurIPS 2023</a> Scholar Award (~$1700).
    </p>
  </li>
  <li>
    <p>
      [Nov '23] Co-presented 2 papers at <a href="https://sites.google.com/view/corl2023-prl/schedule?authuser=0" class="uline">PRL</a>, <a href="https://openreview.net/group?id=robot-learning.org/CoRL/2023/Workshop/TGR" class="uline">TGR</a>, and <a href="https://openreview.net/group?id=robot-learning.org/CoRL/2023/Workshop/LangRob" class="uline">LangRob</a> workshops at <a href="https://www.corl2023.org/" class="uline">CoRL2023</a> in Atlanta, Georgia.
    </p>
  </li>
  <li>
      <p>
        [Sep '23] <a href="https://arxiv.org/pdf/2306.06093.pdf" class="uline">Hyp-NeRF</a> accepted at NeurIPS 2023. See you in New Orleans, Louisiana</a>.
      </p>
  </li>
  <li>
      <p>
        [Sep '23] Joined <a href="https://lis.csail.mit.edu/people/" class="uline">Massachusetts Institute of Technology</a> as a PhD student in EECS</a>.
      </p>
  </li>
  <!-- <li>
      <p>
        [July '23] Served as a reviewer for <a href="https://s2023.siggraph.org/" class="uline">SIGGRAPH 2023</a>.
      </p>
   </li>
   <li> -->
      <p>
        [May '23] I'll be starting as a research intern at <a href="https://mila.quebec/en/" class="uline">Mila - Quebec Artificial Intelligence Institute</a>, Montreal with professors <a href="https://liampaull.ca/" class="uline">Liam Paull</a> and <a href="http://www.cs.toronto.edu/~florian/" class="uline">Florian Shkurti</a>. I will work on learning representations for 3D robotic manipulation. 
      </p>
    </li>
    <li>
      <p>
        [Apr '23] I'll be joining MIT CSAIL as a PhD student this Fall. I will be a part of the <a href="https://www.csail.mit.edu/research/learning-and-intelligent-systems" class="uline">Learning and Intelligent Systems (LIS)</a> Group with professors <a href="https://www.csail.mit.edu/person/leslie-kaelbling" class="uline">Leslie Pack Kaelbling</a> and <a href="https://people.csail.mit.edu/tlp/" class="uline">Tomas Lozano-Perez</a>.
      </p>
    </li>
    <!-- <li>
      <p>
        [Apr '23] Serving as a reviewer for <a href="https://ieee-iros.org/" class="uline">IROS 2023</a>.
      </p>
    </li> -->
    <li>
      <p>
        [Apr '23] Full page abstract on <a href="http://icmpc.org/icmpcprograms/ICMPC17_2023program.pdf" class="uline">"Uncovering Biases Against Indian Artists"</a> accepted at <a href="https://www.icmpc.org/" class="uline">ICMPC17-APSCOM7</a> for a spoken presentation. Awarded a Travel Grant of ¥30,000 to attend the conference in Tokyo, Japan.
      </p>
    </li>
    <li>
      <p>
        [Mar '23] Awarded a generous travel grant of $2250.00 by <a href="https://www.icra2023.org/" class="uline">ICRA 2023</a> IEEE RAS Travel Grant Committee to attend the premier robotics conference in London, UK from 29<sup>th</sup> May to 2<sup>nd</sup> Jun.
      </p>
    </li>
    <li>
      <p>
        [Mar '23] Invited for a talk at Columbia University - slide deck <a href="https://iiitaphyd-my.sharepoint.com/:p:/g/personal/aditya_ag_research_iiit_ac_in/EfulzHAgBLlImxOxLqOZEJYBqOtnMrXz9pqUfL8plYbfVg?e=5cFHkz" class="uline">here</a>. The talk was organized as part of my graduate visit days to Brown, Columbia, and MIT. 
      </p>
    </liw>
    <li>
      <p>
        [Jan '23] 5 works on <a href="images/projects/inr-v_tmlr2022/rnd_poster.pdf" class="uline">Implicit Video Parameterization</a>, <a href="images/projects/faceoff_wacv2023/rnd_poster.pdf" class="uline">V2V Face-Swapping</a>, <a href="images/projects/moocs-lrt_wacv2023/rnd_poster.pdf" class="uline">MOOCs for Lipreading</a>, <a href="images/projects/scarp_icra2023/rnd_poster.pdf" class="uline">3D Shape Completion</a>, and <a href="images/projects/synergistic_iros2023/rnd_poster.pdf" class="uline">Synergistic Tabletop Manipulation</a> presented at <a href="https://rndshowcase.iiit.ac.in/" class="uline">IIIT Hyderabad's RnD showcase</a>.
      </p>
    </li>
    <li>
      <p>
        [Jan '23] 1 paper accepted at <a href="https://www.icra2023.org/" class="uline">ICRA 2023</a> on <a href="https://bipashasen.github.io/scarp/" class="uline">3D Shape Completion in Arbitrary Poses</a>. Featured as the <a href="https://t.co/sT467zT9wv" class="uline">"Publication of the Week"</a> in "Weekly Robotics".
      </p>
    <li>
      <p>
        [Jan '23] Attending <a href="https://sites.google.com/view/researchweek2023/home" class="uline">Google Reserach Week</a> in Bangalore from 29<sup>th</sup> Jan to 31<sup>st</sup> Jan.
      </p>
    <!-- <li>
      <p>
        [Dec '22] Serving as a reviewer at <a href="https://blog.iclr.cc/2022/12/21/announcing-the-accepted-workshops-at-iclr-2023/" class="uline">Neural Fields Workshop</a> at ICLR 2023 (NF2023).
      </p>
    <li> -->
      <!-- <p>
        [Nov '22] Served as a reviewer at <a href="https://www.icra2023.org/" class="uline">ICRA 2023</a> .
      </p> -->
  
    <li>
      <p>
        [Oct '22] Journal paper on <a href="https://openreview.net/forum?id=aIoEkwc2oB" class="uline">representation space for video-based generative tasks</a> accepted at <a href="https://www.jmlr.org/tmlr/" class="uline">TMLR 2022</a>.
      </p>
    <li>
      <p>
        [Aug '22] Two papers on 
        <a href="https://openaccess.thecvf.com/content/WACV2023/papers/Agarwal_FaceOff_A_Video-to-Video_Face_Swapping_System_WACV_2023_paper.pdf" class="uline">video face swapping</a> 
        and <a href="https://openaccess.thecvf.com/content/WACV2023/papers/Agarwal_Towards_MOOCs_for_Lipreading_Using_Synthetic_Talking_Heads_To_Train_WACV_2023_paper.pdf" class="uline">talking-face generation</a> 
        accepted at <a href="https://wacv2023.thecvf.com/home" class="uline">WACV 2023</a> round 1 (acceptance rate 21.6%).
      </p>
    <li>
      <p>
        [May '22] We were in the News (<a href="https://www.iiit.ac.in/files/media/Sakshi-RRC.jpeg" class="uline">news1</a>, <a href="https://blogs.iiit.ac.in/monthly_news/iiiths-robotics-research-centre-proves-mettle-with-two-prestigious-wins/" class="uline">news2</a>) 
        for winning 3rd place at the <a href="https://www.icra2022.org/program/competitions" class="uline">ICRA 2022 international robotics competition</a> 
        on tabletop rearrangement and planning. Awarded a grant of $1000.00.
      </p>
      <p>
        [Oct '21] 1 paper accepted at <a href="https://www.bmvc2021-virtualconference.com/" class="uline">BMVC</a> 
        on <a href="https://www.bmvc2021-virtualconference.com/conference/papers/paper_1468.html" class="uline">lipreading in a one-shot setting using domain adaptation</a>.
      </p>
    <li>
      <p>
        [Mar '21] I will be joining <a href="https://www.iiit.ac.in/" class="uline">IIIT Hyderabad</a> as an MS by Research student.
      </p>
    <li>
      <p>
        [Nov '20] 1 paper accepted at <a href="http://2021.ieeeslt.org/" class="uline">SLT</a> on building <a href="images/projects/reed_slt2021/reed_paper.pdf" class="uline">multilingual acoustic model for low-resource languages</a>.
      </p>
    <li>
      <p>
        [Sep '17] Completed my Bachelor's degree from <a href="https://pes.edu/" class="uline">PES University</a> in Computer Science. Received Academic Distinction Award for exceptional academic performance.
      </p>
    <li>
      <p>
        [Jan '16] I will be interning at the <a href="https://www.ucalgary.ca/" class="uline">University of Calgary</a> in Summer 2016 fully-funded through the <a href="https://www.mitacs.ca/en/programs/globalink/globalink-research-internship" class="uline">MITACS Globalink Research Award</a>. 
      </p>
  </ul>
</div>

<hr/>

<div id="news">
  <h2>Academic Services</h2>
  <ul>
    <li>
      <p>
        Reviewer for <a href="https://2024.ieee-icra.org/" class="uline">ICRA 2024</a>, <a href="https://cvpr.thecvf.com/" class="uline">CVPR 2024</a>.
      </p>
    </li>
    <li>
      <p>
        Reviewer for <a href="https://s2023.siggraph.org/" class="uline">SIGGRAPH 2023</a>, <a href="https://ieee-iros.org/" class="uline">IROS 2023</a>, <a href="https://iclr.cc/virtual/2023/events/workshop" class="uline">ICLR 2023 workshops</a>, <a href="https://www.icra2023.org/" class="uline">ICRA 2023</a>.
      </p>
    </li>
    <li>
      <p>
        [Aug '22] Coordinator for the <a href="https://cvit.iiit.ac.in/summerschool2022/" class="uline">6th CVIT Summer School</a> on AI.
      </p>
    </li>
    <li>
      <p>
        [Aug '22] Gave a <a href="https://youtu.be/QtSH7dv1CwA?t=5353" class="uline">talk</a> on the challenges in tabletop rearrangement and planning at CVIT Summer School 2022.
      </p>
    </li>
    <li>
      <p>
        [Feb '22] I will be taking month long tutorial sessions in machine learning for faculties across universities in India as part of the CSEDU-ML program conducted jointly by IIIT-H, IIT-H, and IIT-D.
      </p>
    </li>
    <li>
      <p>
        [Aug '21] Coordinator for the <a href="https://cvit.iiit.ac.in/summerschool2021/" class="uline">5th CVIT Summer School</a> 
        on AI and conducted tutorial sessions on self-supervised learning and multimodal learning.
      </p>
    </li>
  </ul>
</div>

<hr />

<div id="news">
  <h2>Professional Achievements</h2>
  <ul>
    <li>
      <p>
        [Aug '17] Winners at the <a href="https://blogs.vmware.com/opensource/2017/08/25/open-source-global-borathon/" class="uline">VMWare Global Relay Opensource Borathon</a> among all participating teams at <a href="https://www.vmware.com/" class="uline">VMWare</a>.
      </p>
    </li>
    <li>
      <p>
        [Mar '20] I was selected as one of the two individuals out of 6,000 employees at Microsoft India to undergo a video shoot for the company's campus hiring program. Available on <a href="https://www.youtube.com/watch?v=-ZEXU20tkFw" class="uline">YouTube</a>.
      </p>
    </li>
    <li>
      <p>
        [Mar '18] My work helped scale the <a href="https://communitytraining.microsoft.com/" class="uline">Microsoft Community Training</a> platform to its first 100K users. The work was covered by several media outlets 
        (<a href="https://news.microsoft.com/en-in/microsoft-project-sangam-swacch-bharat-mission/", class="uline">[1]</a>, 
        <a href="https://www.digitalcreed.in/microsofts-project-sangam-accelerates-indias-swachh-bharat-mission/", class="uline">[2]</a>, 
        <a href="https://egov.eletsonline.com/2019/03/microsoft-project-gives-major-boost-to-swachh-bharat-mission/", class="uline">[3]</a>, 
        <a href="https://swachhindia.ndtv.com/government-launches-a-mobile-app-and-website-to-train-citizens-on-sanitation-waste-management-32353/", class="uline">[4]</a>). 
        I was awarded the "Delight your Customer" Award by Microsoft for my outstanding work.
      </p>
    </li>
    <li>
      <p>
        [Feb '17] My work on building <a href="https://www.microsoft.com/en-us/research/lab/microsoft-research-india/" class="uline">Microsoft Research India's</a> flagship project <a href="https://www.microsoft.com/en-us/research/wp-content/uploads/2012/10/MassivelyEmpoweredClassroomsTechReportv1.pdf" class="uline">Massively Empowered Classroom</a> was deployed by <a href="http://web.mie.ac.mu/" class="uline">Mauritius Institute of Education</a>. 
        It was inaugurated by MD MSR India and Minister of Tertiary Education, Mauritius and was covered by the press 
        (<a href="https://education.govmu.org/Documents/archived/2017/SPEECH%20MIE%20FINAL%2008%20FEB%202017.pdf" class="uline">[1]</a>, 
        <a href="https://ict.io/en/launch-of-the-virtual-campus-of-the-mauritius-institute-of-education/" class="uline">[2]</a>, 
        <a href="https://defimedia.info/mauritius-institute-education-mo-klas-build-network-digital-learning-resources" class="uline">[3]</a>).
      </p>
    </li>
  </ul>
</div>  

<div hidden="hidden">
<script type="text/javascript" id="clustrmaps" src="//clustrmaps.com/map_v2.js?d=HlPcyrzc7e8yFyrAN5zVRB4Q6oQOU5EwTtuv9SXJE7Y&cl=ffffff&w=a"></script>
</div>

<hr>
<p align="right">
<small>Forked and modified from <a href="https://virajprabhu.github.io/">Viraj Prabhu's</a> adaptation of <a href="https://github.com/johno/pixyll">Pixyll</a> theme</a></small></p>