Weakly Supervised Cross-Modal Learning for 4D Radar Scene Flow Estimation

• URL: http://arxiv.org/abs/2605.18507v3
• Date: Thu, 21 May 2026 08:43:54 GMT
• Title: Weakly Supervised Cross-Modal Learning for 4D Radar Scene Flow Estimation
• Abstract summary: We propose a task-specific framework for weakly supervised radar scene flow learning.<n>We exploit off-the-shelf 2D tracking and segmentation algorithms to obtain tracked instance masks.<n>Experiments on the real-world View-of-Delft dataset demonstrate that our method surpasses state-of-the-art cross-modal supervised approaches.
• Score: 10.906975408529895
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Due to the difficulty of obtaining ground-truth data for 4D radar scene flow estimation, previous methods typically rely on either self-supervised losses or cross-modal supervision using 3D LiDAR data, 2D images, and odometry. However, self-supervised approaches often yield suboptimal results due to radar's inherently low-fidelity measurements, while existing cross-modal supervised methods introduce complex multi-task architecture and require costly LiDAR sensors to generate pseudo radar scene flow labels from pretrained 3D tracking models. To overcome these limitations, we propose a task-specific iterative framework for weakly supervised radar scene flow learning, using only images and odometry for auxiliary supervision during training. Specially, we establish two novel instance-aware self-supervised losses by exploiting off-the-shelf 2D tracking and segmentation algorithms to obtain tracked instance masks, which are back-projected into 3D space to provide instance-level semantic guidance; for static regions, we integrate vehicle odometry with radar's intrinsic motion cues to construct a rigid static loss. Extensive experiments on the real-world View-of-Delft (VoD) dataset demonstrate that our method not only surpasses state-of-the-art cross-modal supervised approaches that rely on 3D multi-object tracking on dense LiDAR point clouds but also outperforms existing fully supervised scene flow estimation methods. The code is open-sourced at \href{https://github.com/FuJingyun/IterFlow}{https://github.com/FuJingyun/IterFlow}.

Related papers

• RaLiFlow: Scene Flow Estimation with 4D Radar and LiDAR Point Clouds [10.906975408529895]
  We build a Radar-LiDAR scene flow dataset based on a public real-world automotive dataset.<n>We introduce RaLiFlow, the first joint scene flow learning framework for 4D radar and LiDAR.<n>Our method outperforms existing LiDAR-based and radar-based single-modal methods by a significant margin.
  arXiv  Detail & Related papers  (2025-12-11T07:41:33Z)
• VISC: mmWave Radar Scene Flow Estimation using Pervasive Visual-Inertial Supervision [15.903580198464432]
  Current scene flow estimation methods for mmWave radar are typically supervised by dense point clouds from 3D LiDARs.<n>We propose a drift-free rigid transformation estimator that fuses kinematic model-based ego-motions with neural network-learned results.<n>It provides strong supervision signals to radar-based rigid transformation and infers the scene flow of static points.
  arXiv  Detail & Related papers  (2025-07-05T07:53:51Z)
• Street Gaussians without 3D Object Tracker [84.89933388445185]
  Most existing methods rely on labor-intensive manual labeling of object poses.<n>We propose a stable object tracking module by leveraging associations from 2D deep trackers within a 3D object fusion strategy.<n>We address inevitable tracking errors by further introducing a motion learning strategy in an implicit feature space that autonomously corrects trajectory errors and recovers missed detections.
  arXiv  Detail & Related papers  (2024-12-07T05:49:42Z)
• LISO: Lidar-only Self-Supervised 3D Object Detection [25.420879730860936]
  We introduce a novel self-supervised method to train SOTA lidar object detection networks. It works on unlabeled sequences of lidar point clouds only. It utilizes a SOTA self-supervised lidar scene flow network under the hood to generate, track, and iteratively refine pseudo ground truth.
  arXiv  Detail & Related papers  (2024-03-11T18:02:52Z)
• ODM3D: Alleviating Foreground Sparsity for Semi-Supervised Monocular 3D Object Detection [15.204935788297226]
  ODM3D framework entails cross-modal knowledge distillation at various levels to inject LiDAR-domain knowledge into a monocular detector during training. By identifying foreground sparsity as the main culprit behind existing methods' suboptimal training, we exploit the precise localisation information embedded in LiDAR points. Our method ranks 1st in both KITTI validation and test benchmarks, significantly surpassing all existing monocular methods, supervised or semi-supervised.
  arXiv  Detail & Related papers  (2023-10-28T07:12:09Z)
• Robust 3D Object Detection from LiDAR-Radar Point Clouds via Cross-Modal Feature Augmentation [7.364627166256136]
  This paper presents a novel framework for robust 3D object detection from point clouds via cross-modal hallucination. We introduce multiple alignments on both spatial and feature levels to achieve simultaneous backbone refinement and hallucination generation. Experiments on the View-of-Delft dataset show that our proposed method outperforms the state-of-the-art (SOTA) methods for both radar and LiDAR object detection.
  arXiv  Detail & Related papers  (2023-09-29T15:46:59Z)
• Weakly Supervised Monocular 3D Object Detection using Multi-View Projection and Direction Consistency [78.76508318592552]
  Monocular 3D object detection has become a mainstream approach in automatic driving for its easy application. Most current methods still rely on 3D point cloud data for labeling the ground truths used in the training phase. We propose a new weakly supervised monocular 3D objection detection method, which can train the model with only 2D labels marked on images.
  arXiv  Detail & Related papers  (2023-03-15T15:14:00Z)
• Boosting 3D Object Detection by Simulating Multimodality on Point Clouds [51.87740119160152]
  This paper presents a new approach to boost a single-modality (LiDAR) 3D object detector by teaching it to simulate features and responses that follow a multi-modality (LiDAR-image) detector. The approach needs LiDAR-image data only when training the single-modality detector, and once well-trained, it only needs LiDAR data at inference. Experimental results on the nuScenes dataset show that our approach outperforms all SOTA LiDAR-only 3D detectors.
  arXiv  Detail & Related papers  (2022-06-30T01:44:30Z)
• A Lightweight and Detector-free 3D Single Object Tracker on Point Clouds [50.54083964183614]
  It is non-trivial to perform accurate target-specific detection since the point cloud of objects in raw LiDAR scans is usually sparse and incomplete. We propose DMT, a Detector-free Motion prediction based 3D Tracking network that totally removes the usage of complicated 3D detectors.
  arXiv  Detail & Related papers  (2022-03-08T17:49:07Z)
• Anchor-free 3D Single Stage Detector with Mask-Guided Attention for Point Cloud [79.39041453836793]
  We develop a novel single-stage 3D detector for point clouds in an anchor-free manner. We overcome this by converting the voxel-based sparse 3D feature volumes into the sparse 2D feature maps. We propose an IoU-based detection confidence re-calibration scheme to improve the correlation between the detection confidence score and the accuracy of the bounding box regression.
  arXiv  Detail & Related papers  (2021-08-08T13:42:13Z)
• PLUME: Efficient 3D Object Detection from Stereo Images [95.31278688164646]
  Existing methods tackle the problem in two steps: first depth estimation is performed, a pseudo LiDAR point cloud representation is computed from the depth estimates, and then object detection is performed in 3D space. We propose a model that unifies these two tasks in the same metric space. Our approach achieves state-of-the-art performance on the challenging KITTI benchmark, with significantly reduced inference time compared with existing methods.
  arXiv  Detail & Related papers  (2021-01-17T05:11:38Z)
• SelfVoxeLO: Self-supervised LiDAR Odometry with Voxel-based Deep Neural Networks [81.64530401885476]
  We propose a self-supervised LiDAR odometry method, dubbed SelfVoxeLO, to tackle these two difficulties. Specifically, we propose a 3D convolution network to process the raw LiDAR data directly, which extracts features that better encode the 3D geometric patterns. We evaluate our method's performances on two large-scale datasets, i.e., KITTI and Apollo-SouthBay.
  arXiv  Detail & Related papers  (2020-10-19T09:23:39Z)

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.