Panoramic imagery offers a full 360{\deg} field of view and is increasingly common in consumer devices. However, it introduces non-pinhole distortions that challenge joint pose estimation and 3D reconstruction. Existing feed-forward models, built for perspective cameras, generalize poorly to this setting. We propose PanoVGGT, a permutation-equivariant Transformer framework that jointly predicts camera poses, depth maps, and 3D point clouds from one or multiple panoramas in a single forward pass. The model incorporates spherical-aware positional embeddings and a panorama-specific three-axis SO(3) rotation augmentation, enabling effective geometric reasoning in the spherical domain. To resolve inherent global-frame ambiguity, we further introduce a stochastic anchoring strategy during training. In addition, we contribute PanoCity, a large-scale outdoor panoramic dataset with dense depth and 6-DoF pose annotations. Extensive experiments on PanoCity and standard benchmarks demonstrate that PanoVGGT achieves competitive accuracy, strong robustness, and improved cross-domain generalization. Code and dataset will be released.
@article{arxiv.2603.17571,
title = {PanoVGGT: Feed-Forward 3D Reconstruction from Panoramic Imagery},
author = {Yijing Guo and Mengjun Chao and Luo Wang and Tianyang Zhao and Haizhao Dai and Yingliang Zhang and Jingyi Yu and Yujiao Shi},
journal= {arXiv preprint arXiv:2603.17571},
year = {2026}
}