compress images

Tencent · Jun 28, 2024 · d1cd3df · d1cd3df
1 parent 6ba7300
commit d1cd3df
Show file tree

Hide file tree

Showing 25 changed files with 3 additions and 316 deletions.
diff --git a/LICENSE b/LICENSE
diff --git a/assets/images/experiments/cmp1.jpg b/assets/images/experiments/cmp1.jpg
diff --git a/assets/images/experiments/cmp_tm1.jpg b/assets/images/experiments/cmp_tm1.jpg
diff --git a/assets/images/experiments/handaug.jpg b/assets/images/experiments/handaug.jpg
diff --git a/assets/images/experiments/multidiff.jpg b/assets/images/experiments/multidiff.jpg
diff --git a/assets/images/experiments/posealpha.jpg b/assets/images/experiments/posealpha.jpg
diff --git a/assets/images/fig1.png b/assets/images/fig1.png
diff --git a/assets/images/fig4.png b/assets/images/fig4.png
diff --git a/assets/images/pose_conf_method.jpg b/assets/images/pose_conf_method.jpg
diff --git a/assets/showcases/sample_1.png b/assets/showcases/sample_1.png
diff --git a/assets/showcases/sample_2.png b/assets/showcases/sample_2.png
diff --git a/assets/showcases/sample_3.png b/assets/showcases/sample_3.png
diff --git a/assets/showcases/sample_fb_0.png b/assets/showcases/sample_fb_0.png
diff --git a/assets/showcases/sample_fb_1.png b/assets/showcases/sample_fb_1.png
diff --git a/assets/showcases/sample_fb_2.png b/assets/showcases/sample_fb_2.png
diff --git a/assets/showcases/sample_fb_3.png b/assets/showcases/sample_fb_3.png
diff --git a/assets/showcases/sample_hb_0.png b/assets/showcases/sample_hb_0.png
diff --git a/assets/showcases/sample_hb_1.png b/assets/showcases/sample_hb_1.png
diff --git a/assets/showcases/sample_hb_2.png b/assets/showcases/sample_hb_2.png
diff --git a/assets/showcases/sample_hb_3.png b/assets/showcases/sample_hb_3.png
diff --git a/assets/showcases/sample_s0.jpg b/assets/showcases/sample_s0.jpg
diff --git a/assets/showcases/sample_s1.jpg b/assets/showcases/sample_s1.jpg
diff --git a/assets/showcases/sample_s2.jpg b/assets/showcases/sample_s2.jpg
diff --git a/assets/showcases/sample_s3.jpg b/assets/showcases/sample_s3.jpg
diff --git a/index.html b/index.html
@@ -68,7 +68,7 @@ <h1 class="title is-1 publication-title">
               <span class="author-block">
                 <a href="#" target="_blank">Jiaxi Gu</a><sup>1</sup>,</span>
               <span class="author-block">
-                <a href="#" target="_blank">Liwen Wang</a><sup>1</sup>,</span>
+                <a href="#" target="_blank">Li-Wen Wang</a><sup>1</sup>,</span>
               <span class="author-block">
                 <a href="#" target="_blank">Han Wang</a><sup>1,2</sup>,</span>
               <span class="author-block">
@@ -667,7 +667,7 @@ <h2 class="subtitle has-text-centered">
         <h3 class="title is-4">Gradually latent fusion for temporal smoothness</h3>
         <p>
           We propose a progressive approach for generating long video with temporal smoothness.
-          During each denoising step, video clips are firstly denoised separately with the trained model, conditioning on the same reference image and the corresponding sub-sequence of poses. Within each denoising step, the overlapped frames, marked within dashed boxes in the figure, are progressively fused according to their frame positions. 
+          During each denoising step, video segments are firstly denoised separately with the trained model, conditioning on the same reference image and the corresponding sub-sequence of poses. Within each denoising step, the overlapped frames, marked within dashed boxes in the figure, are progressively fused according to their frame positions. 
         </p>
       </div>
       <div class="columns is-centered">
@@ -804,7 +804,7 @@ <h2 class="subtitle has-text-centered">
       <div class="content has-text-justified">
         <h3 class="title is-4">Progressive latent fusion</h3>
         <p>
-          Progressive latent fusion enables smooth transitions and avoids abrupt changes across clip
+          Progressive latent fusion enables smooth transitions and avoids abrupt changes across video segment
           boundaries, thereby enhancing overall visual temporal coherence for long video generation.
         </p>
       </div>