Denoise and Align: Towards Source-Free UDA for Robust Panoramic Semantic Segmentation

TL;DR

This paper introduces DAPASS, a source-free unsupervised domain adaptation framework for panoramic semantic segmentation that effectively handles geometric distortions and class imbalance, achieving state-of-the-art results.

Contribution

The paper proposes DAPASS, a novel source-free UDA method with modules for denoising pseudo-labels and aligning features across scales, addressing privacy constraints and geometric challenges.

Findings

Achieves 55.04% mIoU on Cityscapes-to-DensePASS

Achieves 70.38% mIoU on Stanford2D3D

Outperforms existing methods in source-free panoramic segmentation

Abstract

Panoramic semantic segmentation is pivotal for comprehensive 360{\deg} scene understanding in critical applications like autonomous driving and virtual reality. However, progress in this domain is constrained by two key challenges: the severe geometric distortions inherent in panoramic projections and the prohibitive cost of dense annotation. While Unsupervised Domain Adaptation (UDA) from label-rich pinhole-camera datasets offers a viable alternative, many real-world tasks impose a stricter source-free (SFUDA) constraint where source data is inaccessible for privacy or proprietary reasons. This constraint significantly amplifies the core problems of domain shift, leading to unreliable pseudo-labels and dramatic performance degradation, particularly for minority classes. To overcome these limitations, we propose the DAPASS framework. DAPASS introduces two synergistic modules to robustly transfer knowledge without source data. First, our Panoramic Confidence-Guided Denoising (PCGD) module generates high-fidelity, class-balanced pseudo-labels by enforcing perturbation consistency and incorporating neighborhood-level confidence to filter noise. Second, a Contextual Resolution Adversarial Module (CRAM) explicitly addresses scale variance and distortion by adversarially aligning fine-grained details from high-resolution crops with global semantics from low-resolution contexts. DAPASS achieves state-of-the-art performances on outdoor (Cityscapes-to-DensePASS) and indoor (Stanford2D3D) benchmarks, yielding 55.04% (+2.05%) and 70.38% (+1.54%) mIoU, respectively.