README.md

---
task_categories:
- image-classification
---

The datasets in this repository are used in the paper [Point-Cache: Test-time Dynamic and Hierarchical Cache for Robust and Generalizable Point Cloud Analysis](http://arxiv.org/abs/2503.12150).

### Datasets
1. The folder structure of used datasets should be organized as follows.
```sh
    /path/to/Point-Cache
    |----data # placed in the same level as `runners`, `scripts`, etc. 
        |----modelnet_c
        |----sonn_c
            |----obj_bg
            |----obj_only
            |----hardest
        |----modelnet40
        |----scanobjnn
        |----omniobject3d
            |----1024
            |----4096
            |----16384
        |----objaverse_lvis
    |----runners
    |----scripts
    ...
```

2. You can find the download links of the above datasets from our **huggingface dataset repositories** as follows.
    - [Link](https://huggingface.co/datasets/auniquesun/Point-PRC/tree/main/new-3ddg-benchmarks/xset/corruption) for `modelnet_c` and `sonn_c`
    - [Link](https://huggingface.co/datasets/auniquesun/Point-PRC/tree/main/new-3ddg-benchmarks/xset/dg) for `omniobject3d`
    - [Link](https://huggingface.co/datasets/auniquesun/Point-Cache/tree/main) for `modelnet40`, `scanobjnn`, and `objaverse_lvis`