Showcase media (gifs + 720p mp4s) and Video(decode=False) snippets
Browse files
README.md
CHANGED
|
@@ -55,17 +55,21 @@ configs:
|
|
| 55 |
# CounterStrike-1K
|
| 56 |
|
| 57 |
<p align="center">
|
| 58 |
-
<img src="media/
|
| 59 |
</p>
|
| 60 |
|
| 61 |
**1,490 rendered POV-hours · 7,347 synchronized rounds · 73,470 POV clips · 7 maps · 720p + audio**
|
| 62 |
|
| 63 |
-
<
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
|
| 68 |
-
</
|
|
|
|
|
|
|
|
|
|
|
|
|
| 69 |
|
| 70 |
CounterStrike-1K is the first grounded, professional-grade Counter-Strike 2 dataset with **10 synchronized first-person perspectives per round**, captured from professional match demos. It is designed for video world modeling, action-conditioned video prediction, multi-view consistency research, and audio-conditioned learning.
|
| 71 |
|
|
|
|
| 55 |
# CounterStrike-1K
|
| 56 |
|
| 57 |
<p align="center">
|
| 58 |
+
<img src="media/introduction_video_wall_1000_hours.gif" alt="100+ POV video wall — 1,000+ hours of professional play" width="100%"/>
|
| 59 |
</p>
|
| 60 |
|
| 61 |
**1,490 rendered POV-hours · 7,347 synchronized rounds · 73,470 POV clips · 7 maps · 720p + audio**
|
| 62 |
|
| 63 |
+
<h3 align="center">Synchronized 10-POV rounds with per-frame action overlays</h3>
|
| 64 |
+
<p align="center">
|
| 65 |
+
<img src="media/multi_pov_with_actions.gif" alt="10 synchronized POVs with per-frame action HUD overlays" width="100%"/>
|
| 66 |
+
</p>
|
| 67 |
+
|
| 68 |
+
<h3 align="center">Seven active-duty maps</h3>
|
| 69 |
+
<p align="center">
|
| 70 |
+
<img src="media/seven_maps.gif" alt="Ancient · Anubis · Dust2 · Inferno · Mirage · Nuke · Overpass" width="100%"/><br/>
|
| 71 |
+
<sub>Ancient · Anubis · Dust2 · Inferno · Mirage · Nuke · Overpass</sub>
|
| 72 |
+
</p>
|
| 73 |
|
| 74 |
CounterStrike-1K is the first grounded, professional-grade Counter-Strike 2 dataset with **10 synchronized first-person perspectives per round**, captured from professional match demos. It is designed for video world modeling, action-conditioned video prediction, multi-view consistency research, and audio-conditioned learning.
|
| 75 |
|