arrange seq

index.html

@@ -105,14 +105,14 @@ <h1 class="title is-1 publication-title">STAG4D: Spatial-Temporal Anchored Gener
             <span class="author-block">
               <a href="https://cite.nju.edu.cn/People/Faculty/20190621/i5054.html">Xun Cao</a><sup>1</sup>,
               <span class="author-block">
-                <a href="https://yoyo000.github.io/">Yao Yao</a><sup>1</sup>,</span>
+                <a href="https://yoyo000.github.io/">Yao Yao</a><sup>1✉</sup> </span>
             </span>
           </div>
 
           <div class="is-size-5 publication-authors">
-            <span class="author-block"><sup>1</sup>Nanjing University </span>
-            <span class="author-block"><sup>2</sup>CASIA </span>
-            <span class="author-block"><sup>3</sup>Fudan University </span>
+            <span class="author-block"><sup>1</sup>Nanjing University, </span>
+            <span class="author-block"><sup>2</sup>CASIA, </span>
+            <span class="author-block"><sup>3</sup>Fudan University, </span>
           </div>
 
           <div class="column has-text-centered">
@@ -170,9 +170,7 @@ <h1 class="title is-1 publication-title">STAG4D: Spatial-Temporal Anchored Gener
   <div class="container is-max-desktop">
     <div class="hero-body">
       <h2 class="subtitle has-text-centered">
-        <span class="dnerf">STAG4D</span> is a novel framework that combines pre-trained diffusion models 
-        with dynamic 3D Gaussian splatting for high-fidelity 4D generation. Our method outperforms prior 4D generation works in rendering quality, spatial-temporal consistency, and generation robustness, setting 
-        a new state-of-the-art for 4D generation from diverse inputs, including text, image, and video. 
+        TL; DR: high-fidelity 4D generation from diverse inputs (text, image, and video) with pre-trained diffusion models and dynamic 3D Gaussian splatting.
       </h2>
     </div>
   </div>
@@ -244,6 +242,20 @@ <h2 class="subtitle has-text-centered">
 
 <section class="section">
   <div class="container is-max-desktop">
+    <!--TODO: Paper video. -->
+    <!-- Paper video. -->
+    <div class="columns is-centered has-text-centered">
+      <div class="column is-four-fifths">
+        <h2 class="title is-3">Video</h2>
+        <div class="publication-video">
+          <iframe src="https://www.youtube.com/embed/Gllf19vSF8A"
+                  frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>
+        </div>
+      </div>
+    </div>
+    <!--/ Paper video. -->
+    <!--/ Paper video. -->
+  </div>
     <!-- Abstract. -->
     <div class="columns is-centered has-text-centered">
       <div class="column is-four-fifths">
@@ -266,25 +278,12 @@ <h2 class="title is-3">Abstract</h2>
             Extensive experiments demonstrate that our method outperforms prior 4D generation works in rendering quality, spatial-temporal consistency, and generation robustness, setting a new state-of-the-art for 4D generation from diverse inputs, including text, image, and video. 
 
           </p>
+          <img src="./static/images/pipeline.png">
         </div>
       </div>
     </div>
     <!--/ Abstract. -->
 
-    <!--TODO: Paper video. -->
-    <!-- Paper video. -->
-    <div class="columns is-centered has-text-centered">
-      <div class="column is-four-fifths">
-        <h2 class="title is-3">Video</h2>
-        <div class="publication-video">
-          <iframe src="https://www.youtube.com/embed/Gllf19vSF8A"
-                  frameborder="0" allow="autoplay; encrypted-media" allowfullscreen></iframe>
-        </div>
-      </div>
-    </div>
-    <!--/ Paper video. -->
-    <!--/ Paper video. -->
-  </div>
 </section>
 
 
@@ -373,7 +372,7 @@ <h2 class="title is-3">Text-to-4D</h2>
         <h2 class="title is-3">Application</h2>
 
 
-        <h3 class="title is-4">Multi-view&Normal Generation</h3>
+        <h3 class="title is-4">Multi-view & Normal Generation</h3>
         <div class="content has-text-justified">
           <p>
             Building upon our attention mechanism, we have integrated our design with the newly introduced normal map model by Zero123++.