Improve model card

Hi,

Niels here from the community science team at HF. This PR improves the model card, by
- adding a pipeline tag, so that people find it at https://huggingface.co/models?pipeline_tag=image-feature-extraction
- adding "Usage" section on how to use the models
- link to the paper: https://huggingface.co/papers/2409.04410.

Cheers!

README.md

@@ -2,10 +2,16 @@
 license: apache-2.0
 language:
 - en
+pipeline_tag: image-feature-extraction
 ---
+
 ## Open-MAGVIT2: Democratizing Autoregressive Visual Generation
 
-[[Project Page]](https://github.com/TencentARC/Open-MAGVIT2)
+Code: https://github.com/TencentARC/Open-MAGVIT2
+
+Paper: https://huggingface.co/papers/2409.04410
+
+## Introduction
 
 Until now, VQGAN, the initial tokenizer is still acting an indispensible role in mainstream tasks, especially autoregressive visual generation. Limited by the bottleneck of the size of codebook and the utilization of code, the capability of AR generation with VQGAN is underestimated.
 
@@ -18,3 +24,7 @@ ImageNet 128 × 128:
 
 ImageNet 256 × 256:
 - Model [ImageNet_256_Base.ckpt](https://huggingface.co/TencentARC/Open-MAGVIT2/blob/main/imagenet_256_B.ckpt)
+
+## Usage
+
+Refer to the Github repository which includes [scripts](https://github.com/TencentARC/Open-MAGVIT2/tree/main/scripts) for training, evaluation and inference.