Spatially Optimized Compact Deep Metric Learning Model for Similarity Search

arXiv:2404.06593v1 Announce Type: cross
Abstract: Spatial optimization is often overlooked in many computer vision tasks. Filters should be able to recognize the features of an object regardless of where it is in the image. Similarity search is a crucial task where spatial features decide an important output. The capacity of convolution to capture visual patterns across various locations is limited. In contrast to convolution, the involution kernel is dynamically created at each pixel based on the pixel value and parameters …

abstract arxiv capacity compact computer computer vision cs.ai cs.cv cs.lg features filters image object optimization search spatial tasks type vision