The paper introduces LiDAR‑SAM2, a framework that converts the 2D video foundation model SAM2 into a scalable source of supervision for 4D LiDAR data. By projecting SAM2 video masks into multi‑view LiDAR space and aggregating them temporally, the method automatically generates temporally coherent LiDAR labels without human annotation. Experiments on SemanticKITTI show that these automatically produced semantic and panoptic labels achieve quality close to full human annotation, enabling models trained on them to approach the performance of fully supervised systems.
arXiv:2606. 20189v3 Announce Type: replace-cross Abstract: Leveraging Vision Foundation Models (VFMs) for camera-to-LiDAR knowledge distillation offers a promising solution to the scarcity of annotated data needed to represent the immense geometric and kinematic diversity of real-world autonomous driving (AD).
arXiv:2606. 20189v1 Announce Type: cross Abstract: Leveraging Vision Foundation Models (VFMs) for camera-to-LiDAR knowledge distillation offers a promising solution to the scarcity of annotated data needed to represent the immense geometric and kinematic diversity of real-world autonomous driving (AD).
arXiv:2404.09431v3 Announce Type: replace Abstract: Pseudo-LiDAR has become a promising paradigm for monocular 3D object detection by transforming monocular images into point cloud representations th...
arXiv:2608.21136v1 Announce Type: new Abstract: Recently, open-vocabulary zero-shot 3D scene understanding using vision foundation models has emerged as a promising alternative to data-intensive supe...
arXiv:2607. 00514v1 Announce Type: cross Abstract: Automatic understanding of dynamic 4D point clouds, the 3D-point sequences captured over time by depth sensors and LiDAR, is central to robotics and embodied perception.
Real-world spatial intelligence requires agents to understand scenes from continuous video streams, where objects move, persist, disappear, and reappear over time. While recent spatial foundation models have enabled generalizable feed-forward 3D reconstruction, most streaming methods remain geometry-centric and lack temporally consistent object-level understanding.
LaGen is an autoregressive framework that generates long‑horizon LiDAR scenes frame by frame, using a single‑frame input and bounding‑box conditions to produce high‑fidelity 4D scenes. It introduces a scene decoupling estimation module for better object‑level interaction and a noise modulation module to reduce error accumulation over time. Evaluations on the nuScenes dataset show that LaGen outperforms existing methods, especially on later frames.
arXiv:2607. 10762v1 Announce Type: cross Abstract: Cross-modal distillation from Vision Foundation Models (VFMs) to LiDAR backbones has recently emerged as a self-supervised pretraining strategy that reduces reliance on dense point-wise annotation for 3D scene understanding.
arXiv:2607. 26645v1 Announce Type: cross Abstract: Existing point-based generative methods for outdoor scenes primarily focus on LiDAR-conditioned completion.
Scene coordinate regression (SCR) achieves strong performance in outdoor LiDAR localization, but it usually requires scene-specific training that can take days, limiting practical deployment. Recent w...
arXiv:2501.04001v4 Announce Type: replace Abstract: This work presents Sa2VA, the first comprehensive, unified model for dense grounded understanding of both images and videos. Unlike existing multi-...
arXiv:2512. 09062v2 Announce Type: replace-cross Abstract: Accurate 3D scene interpretation in active construction sites is essential for progress monitoring, safety assessment, and digital twin development.