PanoSAMic: Panoramic Image Segmentation from SAM Feature Encoding and Dual View Fusion

TL;DR

PanoSAMic introduces a novel panoramic image segmentation method that leverages SAM features and dual view fusion, achieving state-of-the-art results on multiple datasets.

Contribution

It adapts the SAM encoder for panoramic images with multi-stage features and a new fusion module for improved segmentation accuracy.

Findings

Achieves SotA results on Stanford2D3DS for multiple modalities.

Outperforms existing methods on Matterport3D for RGB and RGB-D.

Effectively handles distortions and discontinuities in panoramic images.

Abstract

Existing image foundation models are not optimized for spherical images having been trained primarily on perspective images. PanoSAMic integrates the pre-trained Segment Anything (SAM) encoder to make use of its extensive training and integrate it into a semantic segmentation model for panoramic images using multiple modalities. We modify the SAM encoder to output multi-stage features and introduce a novel spatio-modal fusion module that allows the model to select the relevant modalities and best features from each modality for different areas of the input. Furthermore, our semantic decoder uses spherical attention and dual view fusion to overcome the distortions and edge discontinuity often associated with panoramic images. PanoSAMic achieves state-of-the-art (SotA) results on Stanford2D3DS for RGB, RGB-D, and RGB-D-N modalities and on Matterport3D for RGB and RGB-D modalities. https://github.com/dfki-av/PanoSAMic