In this paper, we address panoramic semantic segmentation which is under-explored due to two critical challenges: (1) image distortions and object deformations on panoramas; (2) lack of semantic annotations in the 360{\deg} imagery. To tackle these problems, first, we propose the upgraded Transformer for Panoramic Semantic Segmentation, i.e., Trans4PASS+, equipped with Deformable Patch Embedding (DPE) and Deformable MLP (DMLPv2) modules for handling object deformations and image distortions whenever (before or after adaptation) and wherever (shallow or deep levels). Second, we enhance the Mutual Prototypical Adaptation (MPA) strategy via pseudo-label rectification for unsupervised domain adaptive panoramic segmentation. Third, aside from Pinhole-to-Panoramic (Pin2Pan) adaptation, we create a new dataset (SynPASS) with 9,080 panoramic images, facilitating Synthetic-to-Real (Syn2Real) adaptation scheme in 360{\deg} imagery. Extensive experiments are conducted, which cover indoor and outdoor scenarios, and each of them is investigated with Pin2Pan and Syn2Real regimens. Trans4PASS+ achieves state-of-the-art performances on four domain adaptive panoramic semantic segmentation benchmarks. Code is available at https://github.com/jamycheung/Trans4PASS.
@article{arxiv.2207.11860,
title = {Behind Every Domain There is a Shift: Adapting Distortion-aware Vision Transformers for Panoramic Semantic Segmentation},
author = {Jiaming Zhang and Kailun Yang and Hao Shi and Simon Reiß and Kunyu Peng and Chaoxiang Ma and Haodong Fu and Philip H. S. Torr and Kaiwei Wang and Rainer Stiefelhagen},
journal= {arXiv preprint arXiv:2207.11860},
year = {2024}
}
Comments
Accepted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). Extended version of CVPR 2022 paper arXiv:2203.01452. Code is available at https://github.com/jamycheung/Trans4PASS