Three pillars improving vision foundation model distillation for lidar
Self-supervised image backbones can be used to address complex 2D tasks (eg semantic
segmentation object discovery) very efficiently and with little or no downstream supervision …
segmentation object discovery) very efficiently and with little or no downstream supervision …
Segment any point cloud sequences by distilling vision foundation models
Recent advancements in vision foundation models (VFMs) have opened up new
possibilities for versatile and efficient visual perception. In this work, we introduce Seal, a …
possibilities for versatile and efficient visual perception. In this work, we introduce Seal, a …
Multi-Space Alignments Towards Universal LiDAR Segmentation
A unified and versatile LiDAR segmentation model with strong robustness and
generalizability is desirable for safe autonomous driving perception. This work presents …
generalizability is desirable for safe autonomous driving perception. This work presents …
4d contrastive superflows are dense 3d representation learners
In the realm of autonomous driving, accurate 3D perception is the foundation. However,
develo** such models relies on extensive human annotations–a process that is both …
develo** such models relies on extensive human annotations–a process that is both …
An empirical study of training state-of-the-art LiDAR segmentation models
In the rapidly evolving field of autonomous driving, precise segmentation of LiDAR data is
crucial for understanding complex 3D environments. Traditional approaches often rely on …
crucial for understanding complex 3D environments. Traditional approaches often rely on …
Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation
G Li - Proceedings of the IEEE/CVF Conference on …, 2024 - openaccess.thecvf.com
This paper tackles the domain adaptation problem in point cloud semantic segmentation
which performs adaptation from a fully labeled domain (source domain) to an unlabeled …
which performs adaptation from a fully labeled domain (source domain) to an unlabeled …
Train Till You Drop: Towards Stable and Robust Source-free Unsupervised 3D Domain Adaptation
We tackle the challenging problem of source-free unsupervised domain adaptation (SFUDA)
for 3D semantic segmentation. It amounts to performing domain adaptation on an unlabeled …
for 3D semantic segmentation. It amounts to performing domain adaptation on an unlabeled …
LargeAD: Large-Scale Cross-Sensor Data Pretraining for Autonomous Driving
Recent advancements in vision foundation models (VFMs) have revolutionized visual
perception in 2D, yet their potential for 3D scene understanding, particularly in autonomous …
perception in 2D, yet their potential for 3D scene understanding, particularly in autonomous …
Fusion-then-Distillation: Toward Cross-modal Positive Distillation for Domain Adaptive 3D Semantic Segmentation
In cross-modal unsupervised domain adaptation, a model trained on source-domain data
(eg, synthetic) is adapted to target-domain data (eg, real-world) without access to target …
(eg, synthetic) is adapted to target-domain data (eg, real-world) without access to target …
UniDSeg: Unified Cross-Domain 3D Semantic Segmentation via Visual Foundation Models Prior
3D semantic segmentation using an adapting model trained from a source domain with or
without accessing unlabeled target-domain data is the fundamental task in computer vision …
without accessing unlabeled target-domain data is the fundamental task in computer vision …