Visual Mesh: Real-time Object Detection Using Constant Sample Density
This addresses the problem of computational efficiency in object detection for robotics, offering a significant speed improvement, though it appears incremental as an enhancement of existing convolutional neural networks.
The paper tackled real-time object detection for resource-constrained robotics by proposing Visual Mesh, a geometric input transformation that normalizes pixel and feature density, resulting in execution times sixteen times faster than the fastest competitor while maintaining high accuracy.
This paper proposes an enhancement of convolutional neural networks for object detection in resource-constrained robotics through a geometric input transformation called Visual Mesh. It uses object geometry to create a graph in vision space, reducing computational complexity by normalizing the pixel and feature density of objects. The experiments compare the Visual Mesh with several other fast convolutional neural networks. The results demonstrate execution times sixteen times quicker than the fastest competitor tested, while achieving outstanding accuracy.