Add paper link, project page, and task categories
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,3 +1,19 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
task_categories:
|
| 4 |
+
- video-classification
|
| 5 |
+
- image-to-video
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# Seeing Fast and Slow: Learning the Flow of Time in Videos
|
| 9 |
+
|
| 10 |
+
This repository contains the dataset presented in the paper [Seeing Fast and Slow: Learning the Flow of Time in Videos](https://huggingface.co/papers/2604.21931).
|
| 11 |
+
|
| 12 |
+
[Project Page](https://seeing-fast-and-slow.github.io/)
|
| 13 |
+
|
| 14 |
+
The dataset features a large-scale collection of slow-motion video footage curated from in-the-wild sources. These videos, typically captured by high-speed cameras, contain significantly richer temporal detail than standard footage.
|
| 15 |
+
|
| 16 |
+
The dataset is designed to support research in reasoning about and manipulating the flow of time, including tasks such as:
|
| 17 |
+
- **Speed change detection and playback speed estimation**: Perceiving the passage of time.
|
| 18 |
+
- **Speed-conditioned video generation**: Producing motion at specified playback speeds.
|
| 19 |
+
- **Temporal super-resolution**: Transforming low-FPS sequences into high-FPS sequences with fine-grained temporal details.
|