Rotated object detection in aerial images has received increasing attention for a wide range of applications. However, it is also challenging task due to the huge variations scale, rotation, aspect ratio, and densely arranged targets. Most existing methods heavily rely on large number predefined anchors with different scales, angles, ratios, are optimized distance loss. Therefore, these sensiti...

Anchor-free detectors basically formulate object detection as dense classification and regression. For popular anchor-free detectors, it is common to introduce an individual prediction branch estimate the quality of localization. The following inconsistencies are observed when we delve into practices estimation. Firstly, for some adjacent samples which assigned completely different labels, trai...

Characterized by complicated backgrounds, various types, large size variations, and arbitrary orientations, the detection recognition of arbitrary-oriented objects in remote sensing images are challenging. To address aforementioned problem, an anchor-free object detector using box boundary-aware vectors is proposed. With idea CenterNet to detect as points, oriented achieved predicting center, v...

Using the loop equations we find an explicit expression for genus 1 correction in hermitian two-matrix model in terms of holomorphic objects associated to spectral curve arising in large N limit. Our result generalises known expression for F 1 in hermitian one-matrix model. We discuss the relationship between F 1 , Bergmann tau-function on Hurwitz spaces, G-function of Frobenius manifolds and d...