SeRP: Self-Supervised Representation Learning Using Perturbed Point Clouds
This work addresses self-supervised representation learning for 3D point clouds, which is an incremental improvement for computer vision and robotics applications.
The authors tackled the problem of self-supervised learning for 3D point clouds by proposing SeRP, a framework that uses perturbed inputs and encoder-decoder architectures to reconstruct original point clouds, achieving 0.5-1% higher classification accuracy on ModelNet40 compared to networks trained from scratch.
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.