Datasets:

Modalities:
Video
Size:
< 1K
ArXiv:
Libraries:
Datasets
License:
Harold328 commited on
Commit
e8f3d88
·
verified ·
1 Parent(s): b4bee45

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +24 -3
README.md CHANGED
@@ -1,3 +1,24 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ ---
4
+ # OmniBench-99
5
+
6
+ ## Overview
7
+
8
+ OmniBench-99 benchmark is published in [OmniCreator](https://haroldchen19.github.io/OmniCreator-Page/), containing 99 videos with varied contents (*i.e.*, Environment, Human/Animal, and Object), designed to offer a comprehensive platform
9
+ for evaluating generative video editing, focusing on both editing **types** and **scenarios**.
10
+
11
+ [Paper Link](https://haroldchen19.github.io/OmniCreator-Page/)
12
+
13
+ [Project Page](https://haroldchen19.github.io/OmniCreator-Page/)
14
+
15
+
16
+ ## Dataset Structure
17
+
18
+ Unlike previous benchmarks that evaluate only four editing types, **OmniBench-99** expands the scope to include both editing types and scenarios. Specifically:
19
+
20
+ * *Environment*: Scenarios are developed for **Background**, **Weather**, and **Time** edits.
21
+ * *Object*: Scenarios are created for **Addition**, **Removal**, and **Replacement** edits.
22
+ * *Human/Animal*: Scenarios are designed for **Appearance** and **Motion/Pose** edits.
23
+
24
+