Object Recognition in High-Resolution 360° Panoramic Images Using Spherical Grids

This paper presents an innovative method for object recognition in high-resolution $360^{\circ}$ panoramic images using a VGGl6 convolutional neural network pre-trained on ImageNet. By employing a spherical grid, the method segments panoramic images into sub-images, enabling precise classification of objects such as people and vehicles. After classification, a reconstruction process is used to reassemble the $360^{\circ}$ image, preserving its panoramic properties. Experimental results demonstrate a 98 % accuracy in object recognition, with a mean squared error (MSE) of 0.0096. This method shows potential for object labeling and recognition in ultra-high-resolution images (SK, 8K, IlK), with applications in immersive computer vision, particularly using $360^{\circ}$ camera technology.

I. Introduction

360-degree cameras are increasingly important for capturing immersive, high-resolution imagery in fields like marketing, virtual reality, and social media. Emerging technologies such as Gaussian Splatting [1] and 360 camera-based virtual en-vironment reconstruction [2] are enhancing visual quality, enabling the creation of highly realistic virtual spaces.