mlsanityData to 7 Oct 2026

SAP: Segment Any 4K Panorama

arXiv 2603.12759 ↗First posted 13 Mar 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Promptable instance segmentation is widely adopted in embodied and AR systems, yet the performance of foundation models trained on perspective imagery often degrades on 360deg panoramas. In this paper, we introduce Segment Any 4K Panorama (SAP), a foundation model for 4K high-resolution panoramic instance-level segmentation. We reformulate panoramic segmentation as fixed-trajectory perspective video segmentation, decomposing a panorama into overlapping perspective patches sampled along a continuous spherical traversal. This memory-aligned reformulation preserves native 4K resolution while restoring the smooth viewpoint transitions required for stable cross-view propagation. To enable large-scale supervision, we synthesize 183,440 4K-resolution panoramic images with instance segmentation labels using the InfiniGen engine. Trained under this trajectory-aligned paradigm, SAP generalizes effectively to real-world 360deg images, achieving +17.2 zero-shot mIoU gain over vanilla SAM2 of different sizes on real-world 4K panorama benchmark.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 35 of its 42 references that are in the map. First posted on arXiv 13 Mar 2026.

Builds on

35 of 42 references.

References · 35

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  2. 2023Segment Anythingcited by 8,158
  3. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  4. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  5. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  6. 2024SAM 2: Segment Anything in Images and Videoscited by 2,784
  7. 2025SAM 3: Segment Anything with Conceptscited by 835
  8. 2018Gibson Env: Real-World Perception for Embodied Agentscited by 477
  9. 2023Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistlescited by 287
  10. 2021Reviving Iterative Training with Mask Guidance for Interactive Segmentationcited by 162
  11. 2024Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generationcited by 129
  12. 2020f-BRS: Rethinking Backpropagating Refinement for Interactive Segmentationcited by 117
23 more
  1. 2022Deep Unsupervised Domain Adaptation: A Review of Recent Advances and Perspectivescited by 98
  2. 2022PanoFormer: Panorama Transformer for Indoor 360 Depth Estimationcited by 94
  3. 2025HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixelscited by 91
  4. 2019Tangent Images for Mitigating Spherical Distortioncited by 79
  5. 2022Bending Reality: Distortion-aware Transformers for Adapting to Panoramic Semantic Segmentationcited by 71
  6. 2019Orientation-aware Semantic Segmentation on Icosahedron Spherescited by 52
  7. 2021Capturing Omni-Range Context for Omnidirectional Segmentationcited by 49
  8. 2022Behind Every Domain There is a Shift: Adapting Distortion-aware Vision Transformers for Panoramic Semantic Segmentationcited by 46
  9. 2023Both Style and Distortion Matter: Dual-Path Unsupervised Domain Adaptation for Panoramic Semantic Segmentationcited by 46
  10. 2018SpherePHD: Applying CNNs on a Spherical PolyHeDron Representation of 360 degree Imagescited by 40
  11. 2023Look at the Neighbor: Distortion-aware Unsupervised Domain Adaptation for Panoramic Semantic Segmentationcited by 37
  12. 2024Semantics, Distortion, and Style Matter: Towards Source-free UDA for Panoramic Segmentationcited by 34
  13. 2019DS-PASS: Detail-Sensitive Panoramic Annular Semantic Segmentation through SwaftNet for Surrounding Sensingcited by 30
  14. 2024360SFUDA++: Towards Source-free UDA for Panoramic Segmentation by Learning Reliable Category Prototypescited by 28
  15. 2024GoodSAM: Bridging Domain and Capacity Gaps via Segment Anything Model for Distortion-aware Panoramic Semantic Segmentationcited by 27
  16. 2025OmniSAM: Omnidirectional Segment Anything Model for UDA in Panoramic Semantic Segmentationcited by 26
  17. 2023SGAT4PASS: Spherical Geometry-Aware Transformer for PAnoramic Semantic Segmentationcited by 25
  18. 2025A Survey of Representation Learning, Optimization Strategies, and Applications for Omnidirectional Visioncited by 23
  19. 2024Open Panoramic Segmentationcited by 17
  20. 2025Sat2City: 3D City Generation from A Single Satellite Image with Cascaded Latent Diffusioncited by 15
  21. 2025PANORAMA: The Rise of Omnidirectional Vision in the Embodied AI Eracited by 12
  22. 2024A General Framework to Boost 3D GS Initialization for Text-to-3D Generation by Lexical Richnesscited by 8
  23. 2026PanoSAMic: Panoramic Image Segmentation from SAM Feature Encoding and Dual View Fusioncited by 4

Built on by

Not cited yet.