Unsupervised Multi-Task Feature Learning on Point Clouds
This work addresses the challenge of unsupervised feature learning for 3D point clouds, which is important for applications in computer vision and robotics, and it shows incremental improvements over prior state-of-the-art methods.
The paper tackles the problem of learning features from point clouds without supervision by introducing a multi-task model that jointly learns point and shape features through clustering, reconstruction, and self-supervised classification, achieving 89.1% accuracy on ModelNet40 classification and 68.2 mIoU on ShapeNet segmentation.
We introduce an unsupervised multi-task model to jointly learn point and shape features on point clouds. We define three unsupervised tasks including clustering, reconstruction, and self-supervised classification to train a multi-scale graph-based encoder. We evaluate our model on shape classification and segmentation benchmarks. The results suggest that it outperforms prior state-of-the-art unsupervised models: In the ModelNet40 classification task, it achieves an accuracy of 89.1% and in ShapeNet segmentation task, it achieves an mIoU of 68.2 and accuracy of 88.6%.