Add metadata, link to paper
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -1,3 +1,15 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: transformers
|
| 4 |
+
pipeline_tag: video-segmentation
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
# GLUS: Global-Local Reasoning Unified into A Single Large Language Model for Video Segmentation
|
| 8 |
+
|
| 9 |
+
This repository contains the model described in the paper [GLUS: Global-Local Reasoning Unified into A Single Large Language Model for Video Segmentation](https://arxiv.org/abs/2504.07962).
|
| 10 |
+
|
| 11 |
+
GLUS utilizes global-local reasoning to combine holistic video understanding with detailed frames understanding for video segmentation. It unifies methods of memory bank, object contrastive learning and key frame selection to tackle the problems of mask inconsistency and object obfuscation.
|
| 12 |
+
|
| 13 |
+
Project page: https://glus-video.github.io/
|
| 14 |
+
|
| 15 |
+
Code: https://github.com/GLUS-video/GLUS
|