Spaces:
Runtime error
Runtime error
nit
Browse files- index.html +5 -4
index.html
CHANGED
|
@@ -4,7 +4,7 @@
|
|
| 4 |
<head>
|
| 5 |
<meta charset="utf-8" />
|
| 6 |
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
|
| 7 |
-
<title>
|
| 8 |
<link rel="stylesheet" href="style.css" />
|
| 9 |
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600&display=swap" rel="stylesheet">
|
| 10 |
</head>
|
|
@@ -12,16 +12,17 @@
|
|
| 12 |
<body>
|
| 13 |
<div class="container">
|
| 14 |
<header>
|
| 15 |
-
<h1>V-JEPA 2
|
| 16 |
<p class="subtitle">Real-time video classification using state-of-the-art V-JEPA 2 model pretrained on
|
| 17 |
Something-Something-V2 dataset</p>
|
| 18 |
</header>
|
| 19 |
|
| 20 |
<section class="description">
|
| 21 |
<h2>About the Project</h2>
|
| 22 |
-
<p>This demo showcases the
|
| 23 |
-
state-of-the-art video understanding model that can
|
| 24 |
making it ideal for applications requiring immediate video content analysis.</p>
|
|
|
|
| 25 |
</section>
|
| 26 |
|
| 27 |
<section class="demo">
|
|
|
|
| 4 |
<head>
|
| 5 |
<meta charset="utf-8" />
|
| 6 |
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
|
| 7 |
+
<title>V-JEPA 2 Video Classification Demo</title>
|
| 8 |
<link rel="stylesheet" href="style.css" />
|
| 9 |
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600&display=swap" rel="stylesheet">
|
| 10 |
</head>
|
|
|
|
| 12 |
<body>
|
| 13 |
<div class="container">
|
| 14 |
<header>
|
| 15 |
+
<h1>V-JEPA 2: Streaming Video Classification</h1>
|
| 16 |
<p class="subtitle">Real-time video classification using state-of-the-art V-JEPA 2 model pretrained on
|
| 17 |
Something-Something-V2 dataset</p>
|
| 18 |
</header>
|
| 19 |
|
| 20 |
<section class="description">
|
| 21 |
<h2>About the Project</h2>
|
| 22 |
+
<p>This demo showcases the V-JEPA 2 model's capabilities in real-time video classification. V-JEPA 2 is a
|
| 23 |
+
state-of-the-art video understanding model that can extract video features,
|
| 24 |
making it ideal for applications requiring immediate video content analysis.</p>
|
| 25 |
+
<p>The model is pretrained on Something-Something-V2 dataset and can classify 174 actions.</p>
|
| 26 |
</section>
|
| 27 |
|
| 28 |
<section class="demo">
|