Li, Zhenyu, et al. “SimIPU: Simple 2D Image and 3D Point Cloud Unsupervised Pre-Training for Spatial-Aware Visual Representations”. Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 2, June 2022, pp. 1500-8, doi:10.1609/aaai.v36i2.20040.