SphereSOD improves 360 degree image object detection accuracy and efficiency
SphereSOD: Geometry-Structure Coupled Learning for 360 Salient Object Detection
Computer Vision and Pattern Recognition
Summary
Detecting important objects in 360-degree images is hard because showing the full scene on a flat image causes distortions. The authors created SphereSOD, a new approach that better accounts for the spherical shape of panoramic images when finding important objects. Their method uses geometry to guide how image features are examined and improved at each step. Tests show that SphereSOD finds objects more accurately and quickly than previous methods working directly on spherical images.
What this means in practice
- •For 360-degree video developers: Accurately identify and segment important objects in panoramic videos for enhanced viewer experience and editing tools.
- •For autonomous vehicle engineers: Improve detection of key objects from panoramic sensor data by handling geometric distortions in 360-degree environment perception.
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.