Continual learning on 3D point clouds with random compressed rehearsal

• URL: http://arxiv.org/abs/2205.08013v1
• Date: Mon, 16 May 2022 22:59:52 GMT
• Title: Continual learning on 3D point clouds with random compressed rehearsal
• Authors: Maciej Zamorski, Micha{\l} Stypu{\l}kowski, Konrad Karanowski, Tomasz Trzci\'nski, Maciej Zi\k{e}ba
• Abstract summary: This work proposes a novel neural network architecture capable of continual learning on 3D point cloud data. We utilize point cloud structure properties for preserving a heavily compressed set of past data.
• Score: 10.667104977730304
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Contemporary deep neural networks offer state-of-the-art results when applied to visual reasoning, e.g., in the context of 3D point cloud data. Point clouds are important datatype for precise modeling of three-dimensional environments, but effective processing of this type of data proves to be challenging. In the world of large, heavily-parameterized network architectures and continuously-streamed data, there is an increasing need for machine learning models that can be trained on additional data. Unfortunately, currently available models cannot fully leverage training on additional data without losing their past knowledge. Combating this phenomenon, called catastrophic forgetting, is one of the main objectives of continual learning. Continual learning for deep neural networks has been an active field of research, primarily in 2D computer vision, natural language processing, reinforcement learning, and robotics. However, in 3D computer vision, there are hardly any continual learning solutions specifically designed to take advantage of point cloud structure. This work proposes a novel neural network architecture capable of continual learning on 3D point cloud data. We utilize point cloud structure properties for preserving a heavily compressed set of past data. By using rehearsal and reconstruction as regularization methods of the learning process, our approach achieves a significant decrease of catastrophic forgetting compared to the existing solutions on several most popular point cloud datasets considering two continual learning settings: when a task is known beforehand, and in the challenging scenario of when task information is unknown to the model.

Related papers

• Deep Learning-based 3D Point Cloud Classification: A Systematic Survey and Outlook [12.014972829130764]
  This paper introduces point cloud acquisition, characteristics, and challenges. We review 3D data representations, storage formats, and commonly used datasets for point cloud classification.
  arXiv  Detail & Related papers  (2023-11-05T09:28:43Z)
• AutoSynth: Learning to Generate 3D Training Data for Object Point Cloud Registration [69.21282992341007]
  Auto Synth automatically generates 3D training data for point cloud registration. We replace the point cloud registration network with a much smaller surrogate network, leading to a $4056.43$ speedup. Our results on TUD-L, LINEMOD and Occluded-LINEMOD evidence that a neural network trained on our searched dataset yields consistently better performance than the same one trained on the widely used ModelNet40 dataset.
  arXiv  Detail & Related papers  (2023-09-20T09:29:44Z)
• Explore In-Context Learning for 3D Point Cloud Understanding [71.20912026561484]
  We introduce a novel framework, named Point-In-Context, designed especially for in-context learning in 3D point clouds. We propose the Joint Sampling module, carefully designed to work in tandem with the general point sampling operator. We conduct extensive experiments to validate the versatility and adaptability of our proposed methods in handling a wide range of tasks.
  arXiv  Detail & Related papers  (2023-06-14T17:53:21Z)
• A Survey of Label-Efficient Deep Learning for 3D Point Clouds [109.07889215814589]
  This paper presents the first comprehensive survey of label-efficient learning of point clouds. We propose a taxonomy that organizes label-efficient learning methods based on the data prerequisites provided by different types of labels. For each approach, we outline the problem setup and provide an extensive literature review that showcases relevant progress and challenges.
  arXiv  Detail & Related papers  (2023-05-31T12:54:51Z)
• Self-supervised Learning for Pre-Training 3D Point Clouds: A Survey [25.51613543480276]
  Self-supervised point cloud representation learning has attracted increasing attention in recent years. This paper presents a comprehensive survey of self-supervised point cloud representation learning using DNNs.
  arXiv  Detail & Related papers  (2023-05-08T13:20:55Z)
• Self-Supervised Learning with Multi-View Rendering for 3D Point Cloud Analysis [33.31864436614945]
  We propose a novel pre-training method for 3D point cloud models. Our pre-training is self-supervised by a local pixel/point level correspondence loss and a global image/point cloud level loss. These improved models outperform existing state-of-the-art methods on various datasets and downstream tasks.
  arXiv  Detail & Related papers  (2022-10-28T05:23:03Z)
• Learning-based Point Cloud Registration for 6D Object Pose Estimation in the Real World [55.7340077183072]
  We tackle the task of estimating the 6D pose of an object from point cloud data. Recent learning-based approaches to addressing this task have shown great success on synthetic datasets. We analyze the causes of these failures, which we trace back to the difference between the feature distributions of the source and target point clouds.
  arXiv  Detail & Related papers  (2022-03-29T07:55:04Z)
• Unsupervised Point Cloud Representation Learning with Deep Neural Networks: A Survey [104.71816962689296]
  Unsupervised point cloud representation learning has attracted increasing attention due to the constraint in large-scale point cloud labelling. This paper provides a comprehensive review of unsupervised point cloud representation learning using deep neural networks.
  arXiv  Detail & Related papers  (2022-02-28T07:46:05Z)
• Review: deep learning on 3D point clouds [9.73176900969663]
  Point cloud is one of the most significant data formats for 3D representation. Deep learning is now the most powerful tool for data processing in computer vision.
  arXiv  Detail & Related papers  (2020-01-17T12:55:23Z)
• Deep Learning for 3D Point Clouds: A Survey [58.954684611055]
  This paper presents a review of recent progress in deep learning methods for point clouds. It covers three major tasks, including 3D shape classification, 3D object detection and tracking, and 3D point cloud segmentation. It also presents comparative results on several publicly available datasets.
  arXiv  Detail & Related papers  (2019-12-27T09:15:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.