6 papers
Is Contrastive Distillation Enough for Learning Comprehensive 3D Representations?
Yifan Zhang, Junhui Hou
Cross-modal contrastive distillation has recently been explored for learning effective 3D representations. However, existing methods focus primarily on modality-shared features, ne…
Unsupervised Online 3D Instance Segmentation with Synthetic Sequences and Dynamic Loss
Yifan Zhang, Wei Zhang, Chuangxin He +2
Unsupervised online 3D instance segmentation is a fundamental yet challenging task, as it requires maintaining consistent object identities across LiDAR scans without relying on an…
Optimizing Multi-Modality Trackers via Significance-Regularized Tuning
Zhiwen Chen, Jinjian Wu, Zhiyu Zhu +3
This paper tackles the critical challenge of optimizing multi-modality trackers by effectively adapting pre-trained models for RGB data. Existing fine-tuning paradigms oscillate be…
Reflectance Prediction-based Knowledge Distillation for Robust 3D Object Detection in Compressed Point Clouds
Hao Jing, Anhong Wang, Yifan Zhang +2
Regarding intelligent transportation systems, low-bitrate transmission via lossy point cloud compression is vital for facilitating real-time collaborative perception among connecte…
Self-supervised Learning of LiDAR 3D Point Clouds via 2D-3D Neural Calibration
Yifan Zhang, Junhui Hou, Siyu Ren +3
This paper introduces a novel self-supervised learning framework for enhancing 3D perception in autonomous driving scenes. Specifically, our approach, namely NCLR, focuses on 2D-3D…
Fine-grained Image-to-LiDAR Contrastive Distillation with Visual Foundation Models
Yifan Zhang, Junhui Hou
Contrastive image-to-LiDAR knowledge transfer, commonly used for learning 3D representations with synchronized images and point clouds, often faces a self-conflict dilemma. This is…