360 degree object detection improves by linking spherical shape and object features
SphereSOD: Geometry-Structure Coupled Learning for 360 Salient Object Detection
Computer Vision and Pattern Recognition
Summary
Detecting important objects in images that capture everything around, like a globe, is tricky because when the round image is shown flat, it gets stretched and distorted. The authors made a system called SphereSOD that pays close attention to how the round shape bends and how object outlines appear to get better at finding these important objects. Their method uses the natural sphere shape to guide how it looks for details and refines its guesses in steps. Tests show their system works well and fast compared to others, handling the distortion without changing the image projection too much.
360 degree imagessalient object detectionequirectangular projectionspherical geometryfeature samplingcontext aggregationimage segmentationprojection distortiondeformable samplingprogressive decoding
Authors
Junsong Zhang, Zhijie Shen, Shuai Zheng, Feng Li, Runmin Cong, Yao Zhao, Chunyu Lin
Abstract
360° salient object detection (SOD) aims to accurately segment salient regions across a full field of view. However, equirectangular projection (ERP) introduces severe spatial distortion when mapping the spherical domain onto a planar representation. Existing methods mainly focus on compensating projection distortion while overlooking the interaction between panoramic geometry and salient object structure during feature perception and prediction refinement. To this end, we propose SphereSOD, an ERP-native framework that couples panoramic geometry with evolving salient structures. Specifically, spherical geometry governs feature sampling and spatial weighting, while coarse-grained saliency and contour prediction influence context aggregation during the progressive decoding process. SphereSOD first initializes deformable sampling based on spherical projection geometry and then employs bounded, content-adaptive offsets, yielding features that are better aligned with the underlying panoramic geometry. Subsequently, the decoder performs structure-guided context aggregation and progressive refinement to recover complete salient regions and accurate boundaries. Extensive experiments on three public 360° SOD benchmarks demonstrate state-of-the-art performance and a favorable accuracy-efficiency trade-off, supporting structurepreserving inference directly in ERP space as a promising alternative to projection-heavy panoramic pipelines.