What's in my Room? Object Recognition on Indoor Panoramic Images
This work addresses the challenging problem of object recognition in indoor panoramic images, which has not been deeply investigated, for applications in scene understanding and robotics.
The paper tackles object recognition in indoor panoramic images by developing a system that performs object detection and semantic segmentation, then transforms results into 3D bounding boxes placed in room models. It reports that the method outperforms state-of-the-art approaches by a large margin, demonstrating a complete understanding of main objects in indoor scenes.
In the last few years, there has been a growing interest in taking advantage of the 360 panoramic images potential, while managing the new challenges they imply. While several tasks have been improved thanks to the contextual information these images offer, object recognition in indoor scenes still remains a challenging problem that has not been deeply investigated. This paper provides an object recognition system that performs object detection and semantic segmentation tasks by using a deep learning model adapted to match the nature of equirectangular images. From these results, instance segmentation masks are recovered, refined and transformed into 3D bounding boxes that are placed into the 3D model of the room. Quantitative and qualitative results support that our method outperforms the state of the art by a large margin and show a complete understanding of the main objects in indoor scenes.