Rethinking LiDAR Domain Generalization: Single Source as Multiple Density Domains

• URL: http://arxiv.org/abs/2312.12098v2
• Date: Tue, 16 Jul 2024 11:24:01 GMT
• Title: Rethinking LiDAR Domain Generalization: Single Source as Multiple Density Domains
• Authors: Jaeyeul Kim, Jungwan Woo, Jeonghoon Kim, Sunghoon Im,
• Abstract summary: We propose a Density Discriminative Feature Embedding (DDFE) module for domain generalization. The DDFE module is meticulously designed to extract density-specific features within a single source domain. In addition, we introduce a simple yet effective density augmentation technique aimed at expanding the spectrum of density in source data.
• Score: 11.26654706501546
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In the realm of LiDAR-based perception, significant strides have been made, yet domain generalization remains a substantial challenge. The performance often deteriorates when models are applied to unfamiliar datasets with different LiDAR sensors or deployed in new environments, primarily due to variations in point cloud density distributions. To tackle this challenge, we propose a Density Discriminative Feature Embedding (DDFE) module, capitalizing on the observation that a single source LiDAR point cloud encompasses a spectrum of densities. The DDFE module is meticulously designed to extract density-specific features within a single source domain, facilitating the recognition of objects sharing similar density characteristics across different LiDAR sensors. In addition, we introduce a simple yet effective density augmentation technique aimed at expanding the spectrum of density in source data, thereby enhancing the capabilities of the DDFE. Our DDFE stands out as a versatile and lightweight domain generalization module. It can be seamlessly integrated into various 3D backbone networks, where it has demonstrated superior performance over current state-of-the-art domain generalization methods. Code is available at https://github.com/dgist-cvlab/MultiDensityDG.

Related papers

• Real-IAD D3: A Real-World 2D/Pseudo-3D/3D Dataset for Industrial Anomaly Detection [53.2590751089607]
  Real-IAD D3 is a high-precision multimodal dataset that incorporates an additional pseudo3D modality generated through photometric stereo. We introduce an effective approach that integrates RGB, point cloud, and pseudo-3D depth information to leverage the complementary strengths of each modality. Our experiments highlight the importance of these modalities in boosting detection robustness and overall IAD performance.
  arXiv  Detail & Related papers  (2025-04-19T08:05:47Z)
• Improving Generalization Ability for 3D Object Detection by Learning Sparsity-invariant Features [21.761631081209263]
  We propose a method to improve the generalization ability for 3D object detection on a single domain. To learn sparsity-invariant features from a single source domain, we selectively subsample the source data to a specific beam. We also employ the teacher-student framework to align the Bird's Eye View features for different point clouds densities.
  arXiv  Detail & Related papers  (2025-02-04T13:47:02Z)
• DFR-Net: Density Feature Refinement Network for Image Dehazing Utilizing Haze Density Difference [0.6165605009782557]
  In image dehazing task, haze density is a key feature and affects the performance of dehazing methods. We propose a density-aware dehazing method named Density Feature Refinement Network (DFR-Net) that extracts haze density features from density differences.
  arXiv  Detail & Related papers  (2023-07-26T02:53:29Z)
• Semantic Segmentation on 3D Point Clouds with High Density Variations [44.467561618769714]
  HDVNet contains a nested set of encoder-decoder pathways, each handling a specific point density range. By effectively handling input density variations, HDVNet outperforms state-of-the-art models in segmentation accuracy on real point clouds with inconsistent density.
  arXiv  Detail & Related papers  (2023-07-04T05:44:13Z)
• Density-Insensitive Unsupervised Domain Adaption on 3D Object Detection [19.703181080679176]
  3D object detection from point clouds is crucial in safety-critical autonomous driving. We propose a density-insensitive domain adaption framework to address the density-induced domain gap. Experimental results on three widely adopted 3D object detection datasets demonstrate that our proposed domain adaption method outperforms the state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-04-19T06:33:07Z)
• RAIN: RegulArization on Input and Network for Black-Box Domain Adaptation [80.03883315743715]
  Source-free domain adaptation transits the source-trained model towards target domain without exposing the source data. This paradigm is still at risk of data leakage due to adversarial attacks on the source model. We propose a novel approach named RAIN (RegulArization on Input and Network) for Black-Box domain adaptation from both input-level and network-level regularization.
  arXiv  Detail & Related papers  (2022-08-22T18:18:47Z)
• LiDAR Distillation: Bridging the Beam-Induced Domain Gap for 3D Object Detection [96.63947479020631]
  In many real-world applications, the LiDAR points used by mass-produced robots and vehicles usually have fewer beams than that in large-scale public datasets. We propose the LiDAR Distillation to bridge the domain gap induced by different LiDAR beams for 3D object detection.
  arXiv  Detail & Related papers  (2022-03-28T17:59:02Z)
• Point Density-Aware Voxels for LiDAR 3D Object Detection [8.136649838488042]
  Point Density-Aware Voxel network (PDV) is an end-to-end two stage LiDAR 3D object detection architecture. PDV efficiently localizes voxel features from the 3D sparse backbone through voxel point centroids. PDV outperforms all state-of-the-art methods on the Open dataset.
  arXiv  Detail & Related papers  (2022-03-10T22:11:06Z)
• META: Mimicking Embedding via oThers' Aggregation for Generalizable Person Re-identification [68.39849081353704]
  Domain generalizable (DG) person re-identification (ReID) aims to test across unseen domains without access to the target domain data at training time. This paper presents a new approach called Mimicking Embedding via oThers' Aggregation (META) for DG ReID.
  arXiv  Detail & Related papers  (2021-12-16T08:06:50Z)
• Domain Attention Consistency for Multi-Source Domain Adaptation [100.25573559447551]
  Key design is a feature channel attention module, which aims to identify transferable features (attributes) Experiments on three MSDA benchmarks show that our DAC-Net achieves new state of the art performance on all of them.
  arXiv  Detail & Related papers  (2021-11-06T15:56:53Z)
• ePointDA: An End-to-End Simulation-to-Real Domain Adaptation Framework for LiDAR Point Cloud Segmentation [111.56730703473411]
  Training deep neural networks (DNNs) on LiDAR data requires large-scale point-wise annotations. Simulation-to-real domain adaptation (SRDA) trains a DNN using unlimited synthetic data with automatically generated labels. ePointDA consists of three modules: self-supervised dropout noise rendering, statistics-invariant and spatially-adaptive feature alignment, and transferable segmentation learning.
  arXiv  Detail & Related papers  (2020-09-07T23:46:08Z)
• Density-Aware Graph for Deep Semi-Supervised Visual Recognition [102.9484812869054]
  Semi-supervised learning (SSL) has been extensively studied to improve the generalization ability of deep neural networks for visual recognition. This paper proposes to solve the SSL problem by building a novel density-aware graph, based on which the neighborhood information can be easily leveraged.
  arXiv  Detail & Related papers  (2020-03-30T02:52:40Z)

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.