Research
PanoSAMic: Panoramic Image Segmentation from SAM Feature Encoding and Dual View Fusion
arXiv:2601.07447v2 Announce Type: replace Abstract: Existing image foundation models are not optimized for spherical images having been trained primarily on perspective images. PanoSAMic integrates th
arXiv:2601.07447v2 Announce Type: replace 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
Related
- Adapting Foundation Models for Annotation-Efficient Adnexal Mass Segmentation in Cine Images
- Cross-Attentive Multiview Fusion of Vision-Language Embeddings
- Uncertainty-Based Ensemble Learning in CMR Semantic Segmentation
- FDIF: Formula-Driven supervised Learning with Implicit Functions for 3D Medical Image Segmentation
- SwinTextUNet: Integrating CLIP-Based Text Guidance into Swin Transformer U-Nets for Medical Image Segmentation
- M^{2}SNet: Multi-scale in Multi-scale Subtraction Network for Medical Image Segmentation
Source: arXiv cs.CV | 2026-04-14