{
  "id": "2209.06067",
  "version": "v1",
  "published": "2022-09-13T15:22:36.000Z",
  "updated": "2022-09-13T15:22:36.000Z",
  "title": "SeRP: Self-Supervised Representation Learning Using Perturbed Point Clouds",
  "authors": [
    "Siddhant Garg",
    "Mudit Chaudhary"
  ],
  "comment": "6 pages",
  "categories": [
    "cs.CV",
    "cs.AI"
  ],
  "abstract": "We present SeRP, a framework for Self-Supervised Learning of 3D point clouds. SeRP consists of encoder-decoder architecture that takes perturbed or corrupted point clouds as inputs and aims to reconstruct the original point cloud without corruption. The encoder learns the high-level latent representations of the points clouds in a low-dimensional subspace and recovers the original structure. In this work, we have used Transformers and PointNet-based Autoencoders. The proposed framework also addresses some of the limitations of Transformers-based Masked Autoencoders which are prone to leakage of location information and uneven information density. We trained our models on the complete ShapeNet dataset and evaluated them on ModelNet40 as a downstream classification task. We have shown that the pretrained models achieved 0.5-1% higher classification accuracies than the networks trained from scratch. Furthermore, we also proposed VASP: Vector-Quantized Autoencoder for Self-supervised Representation Learning for Point Clouds that employs Vector-Quantization for discrete representation learning for Transformer-based autoencoders.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2022-09-13T15:22:36.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "self-supervised representation learning",
      "perturbed point clouds",
      "autoencoder",
      "original point cloud",
      "high-level latent representations"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 6,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}