Updated with ICCV paper

Author: kevincxie

assets/phys_pe.png

@@ -16,8 +16,8 @@ <h1 class="post-title">Kevin Xie</h1>
   <article class="post-content">
     <!-- {{ content }} -->
 
-    I am a Masters student in the Department of Computer Science at the University of Toronto and the Vector Institute, gratefully co-supervised by Florian Shkurti and Sanja Fidler.
-    I am also a Research Scientist Intern at Nvidia.
+    I am a PhD student in the Department of Computer Science at the University of Toronto and the Vector Institute, gratefully co-supervised by Florian Shkurti and Sanja Fidler.
+    I am also a Research Scientist on the Toronto AI Lab at Nvidia.
     Before that I did my undergrad in Engineering Physics at UBC, where I had the great pleasure of working with Michiel van de Panne and his group.
     <br>
     My research interests surround learning-based methods for simulation and control and include generative modelling, model-based reinforcement learning and character control.
@@ -29,6 +29,35 @@ <h1 class="post-title">Kevin Xie</h1>
   {% include page_divider.html %}
   <h2> Publications</h2>
   <table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
+
+    <tr>
+        <td style="padding:6px;width:33%;vertical-align:middle">
+          <!-- <div style = "height:2.5em"> -->
+            <!-- <img src='assets/PLAiD_teaser.png' class="heightSet"> -->
+            <img class="paper-pic" src='assets/phys_pe.png'>
+            <!-- </div> -->
+        </td>
+        <td style="padding:6px;width:66%;vertical-align:middle">
+          <strong>Physics-based Human Motion Estimation and Synthesis from Videos</strong>
+          <br>
+          <strong>Kevin Xie</strong>,
+          <a>Tingwu Wang</a>,
+          <a>Umar Iqbal</a>,
+          <a>Yunrong Guo</a>,
+          <a>Sanja Fidler</a>
+          <a>Florian Shkurti</a>
+          <!-- <a href="https://arxiv.org/abs/2010.08888"> -->
+          <!-- </a> -->
+          <br>
+            Video pose estimation quality can be improved greatly by enforcing physics consistency.
+          <br>
+          <em>ICCV</em>, 2021
+          <!-- <a href="https://arxiv.org/abs/2010.08888">arXiv</a> -->
+          <a href="https://arxiv.org/abs/2109.09913">paper</a>
+          <a href="https://nv-tlabs.github.io/physics-pose-estimation-project-page/">website</a>
+          <p></p>
+        </td>
+    </tr>
     <tr>
         <td style="padding:6px;width:33%;vertical-align:middle">
           <!-- <div style = "height:2.5em"> -->
@@ -49,7 +78,7 @@ <h2> Publications</h2>
           <br>
             Learning temporally extended skills for planning with a world model improves MBRL training and transfer.
           <br>
-          <em>ICLR</em>, 2021 (to appear)
+          <em>ICLR</em>, 2021
           <!-- <a href="https://arxiv.org/abs/2010.08888">arXiv</a> -->
           <a href="https://arxiv.org/pdf/2011.13897.pdf">paper</a>
           <a href="https://sites.google.com/view/partial-amortization-hierarchy/home">website</a>