Update index.html

index.html

@@ -160,8 +160,8 @@ <h2 class="title is-3">Abstract</h2>
 <!-- Image 1 -->
 <section class="hero is-small">
   <div class="hero-body">
-    <h2 class="title is-3">KeyBot framework</h2>
     <div class="container has-text-centered">
+        <h2 class="title is-3">KeyBot framework</h2>
       <img class="centered-image" src="static/images/fig_flow_new.png" alt="MY ALT TEXT" width="70%"/>
       <p class="subtitle has-text-centered">
         <b>Overview of (a) interactive keypoint estimation and (b) the KeyBot framework.</b> The interaction model generates initial keypoint predictions from an image, which the user then revises, leading to updated results. This approach requires user input for error correction. In contrast, KeyBot offers a cost-efficient feedback mechanism via automated, rapid, and iterative refinement, without needing user input. It independently identifies and corrects major errors, significantly reducing the need for user-based error detection, thus allowing users to concentrate on the final adjustments and verifications.
@@ -174,8 +174,8 @@ <h2 class="title is-3">KeyBot framework</h2>
 <!-- Image 2 -->
 <section class="hero is-light">
   <div class="hero-body">
-    <h2 class="title is-3">Training overview</h2>
     <div class="container has-text-centered">
+      <h2 class="title is-3">Training overview</h2>
       <img class="centered-image" src="static/images/fig_method_new.png" alt="MY ALT TEXT" width="70%"/>
       <p class="subtitle has-text-centered">
         <b>Training overview of KeyBot.</b> KeyBot consists of two main components, the detector and the corrector. (a) Synthetic errors are introduced to the groundtruth keypoints, creating inaccurate keypoints and corresponding anomaly labels. (b) The detector is trained to discern whether each input keypoint is accurate or not. (c) The corrector is trained to refine these inaccurate keypoints accurately.