English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The ability of scene understanding has sparked active research for panoramicimage semantic segmentation. However, the performance is hampered by distortionof the equirectangular projection (ERP) and a lack of pixel-wise annotations.For this reason, some works treat the ERP and pinhole images equally andtransfer knowledge from the pinhole to ERP images via unsupervised domainadaptation (UDA). However, they fail to handle the domain gaps caused by: 1)the inherent differences between camera sensors and captured scenes; 2) thedistinct image formats (e.g., ERP and pinhole images). In this paper, wepropose a novel yet flexible dual-path UDA framework, DPPASS, taking ERP andtangent projection (TP) images as inputs. To reduce the domain gaps, we proposecross-projection and intra-projection training. The cross-projection trainingincludes tangent-wise feature contrastive training and prediction consistencytraining. That is, the former formulates the features with the same projectionlocations as positive examples and vice versa, for the models' awareness ofdistortion, while the latter ensures the consistency of cross-model predictionsbetween the ERP and TP. Moreover, adversarial intra-projection training isproposed to reduce the inherent gap, between the features of the pinhole imagesand those of the ERP and TP images, respectively. Importantly, the TP path canbe freely removed after training, leading to no additional inference cost.Extensive experiments on two benchmarks show that our DPPASS achieves +1.06$\%$mIoU increment than the state-of-the-art approaches.

Both Style and Distortion Matter: Dual-Path Unsupervised Domain Adaptation for Panoramic Semantic Segmentation