HVDistill: Transferring Knowledge from Images to Point Clouds via Unsupervised Hybrid-View Distillation

arXiv:2403.11817v1 Announce Type: new
Abstract: We present a hybrid-view-based knowledge distillation framework, termed HVDistill, to guide the feature learning of a point cloud neural network with a pre-trained image network in an unsupervised man- ner. By exploiting the geometric relationship between RGB cameras and LiDAR sensors, the correspondence between the two modalities based on both image- plane view and bird-eye view can be established, which facilitates representation learning. Specifically, the image-plane correspondences can be simply ob- tained by projecting the …

abstract arxiv cameras cloud cs.cv distillation feature framework guide hybrid image images knowledge lidar ner network neural network relationship sensors type unsupervised via view