Self-Supervised Learning with Multi-View Rendering for 3D Point Cloud Analysis

November 29, 2022

Recently, great progress has been made in 3D deep learning with the emergence of deep neural networks specifically designed for 3D point clouds. These networks are often trained from scratch or from pre-trained models learned purely from point cloud data. Inspired by the success of deep learning in the image domain, we devise a novel pre-training technique for better model initialization by utilizing the multi-view rendering of the 3D data. Our pre-training is self-supervised by a local pixel/point level correspondence loss computed from perspective projection and a global image/point cloud level loss based on knowledge distillation, thus effectively improving upon popular point cloud networks, including PointNet, DGCNN and SR-UNet. These improved models outperform existing state-of-the-art methods on various datasets and downstream tasks. We also analyze the benefits of synthetic and real data for pre-training, and observe that pre-training on synthetic data is also useful for high-level downstream tasks. Code and pre-trained models are available at


< 1 minute

Bach Tran, Binh-Son Hua, Anh Tuan Tran, and Minh Hoai

ACCV 2022

Share Article

Related publications

CV AAAI Top Tier
January 8, 2024

Yifeng*, Duc Nguyen Duy*, Lam Nguyen Thanh, Cuong Pham, Minh Hoai

January 8, 2024

Tran Huynh Ngoc, Dang Minh Nguyen, Tung Pham, Anh Tran

CV NeurIPS Top Tier
October 4, 2023

Quang Nguyen, Vu Tuan Truong, Anh Tran, Khoi Nguyen

CV NeurIPS Top Tier
October 4, 2023

Dung Nguyen, Tuan Nguyen, Anh Tran, Khoa Doan, Kok-seng Wong