Correct pipeline tag and update tags

This PR corrects the `pipeline_tag` to accurately reflect the model's functionality as video-to-3D point cloud generation. The tag is changed from `depth-estimation` to `video-to-3d`. The `video-to-4d` tag is also removed as it's not directly represented in the model's output. A link to the paper is also added.

README.md

@@ -1,14 +1,13 @@
 ---
-library_name: geometry-crafter
-license: other
-language:
-- en
 base_model:
 - tencent/DepthCrafter
 - stabilityai/stable-video-diffusion-img2vid-xt
-pipeline_tag: depth-estimation
+language:
+- en
+library_name: geometry-crafter
+license: other
+pipeline_tag: video-to-3d
 tags:
-- video-to-4d
 - point-cloud
 ---
 
@@ -43,7 +42,7 @@ If you find GeometryCrafter useful, **please help ⭐ this repo**, which is impo
 
 ## 📝 Introduction
 
-We present GeometryCrafter, a novel approach that estimates temporally consistent, high-quality point maps from open-world videos, facilitating downstream applications such as 3D/4D reconstruction and depth-based video editing or generation.
+We present GeometryCrafter, a novel approach that estimates temporally consistent, high-quality point maps from open-world videos, facilitating downstream applications such as 3D/4D reconstruction and depth-based video editing or generation. This model is described in detail in the paper [GeometryCrafter: Consistent Geometry Estimation for Open-world Videos with Diffusion Priors](https://arxiv.org/abs/2504.01016).
 
 Release Notes:
 - `[01/04/2025]` 🔥🔥🔥**GeometryCrafter** is released now, have fun!
@@ -138,3 +137,11 @@ Please check the `evaluation` folder.
 
 - Welcome to open issues and pull requests.
 - Welcome to optimize the inference speed and memory usage, e.g., through model quantization, distillation, or other acceleration techniques.
+
+## 📜 Citation
+
+If you find this work helpful, please consider citing:
+
+```BibTeXw
+  TODO
+```