We introduce JointNet, a novel neural network architecture for modeling the joint distribution of images and an additional dense modality (e.g., depth maps). JointNet is extended from a pre-trained text-to-image diffusion model, where a copy of the original network is created for the new dense modality branch and is densely connected with the RGB branch. The RGB branch is locked during network fine-tuning, which enables efficient learning of the new modality distribution while maintaining the strong generalization ability of the large-scale pre-trained diffusion model. We demonstrate the effectiveness of JointNet by using RGBD diffusion as an example and through extensive experiments, showcasing its applicability in a variety of applications, including joint RGBD generation, dense depth prediction, depth-conditioned image generation, and coherent tile-based 3D panorama generation.
@article{arxiv.2310.06347,
title = {JointNet: Extending Text-to-Image Diffusion for Dense Distribution Modeling},
author = {Jingyang Zhang and Shiwei Li and Yuanxun Lu and Tian Fang and David McKinnon and Yanghai Tsin and Long Quan and Yao Yao},
journal= {arXiv preprint arXiv:2310.06347},
year = {2023}
}