论文标题
在2D对象检测中的双曲线嵌入
On Hyperbolic Embeddings in 2D Object Detection
论文作者
论文摘要
在大多数情况下，对象检测已在欧几里得空间中进行配制，在欧几里得空间中，欧几里得或球形地球距离测量了图像区域与对象类原型的相似性。在这项工作中，我们研究双曲线几何形状是否更好地匹配对象分类空间的基础结构。我们将双曲线分类器组合到两阶段，基于关键点和基于变压器的对象检测体系结构中，并在大规模，长尾和零摄像的对象检测基准上对其进行评估。在我们广泛的实验评估中，我们观察到分类空间结构中出现的分类类别层次结构，从而导致较低的分类错误并提高整体对象检测性能。
Object detection, for the most part, has been formulated in the euclidean space, where euclidean or spherical geodesic distances measure the similarity of an image region to an object class prototype. In this work, we study whether a hyperbolic geometry better matches the underlying structure of the object classification space. We incorporate a hyperbolic classifier in two-stage, keypoint-based, and transformer-based object detection architectures and evaluate them on large-scale, long-tailed, and zero-shot object detection benchmarks. In our extensive experimental evaluations, we observe categorical class hierarchies emerging in the structure of the classification space, resulting in lower classification errors and boosting the overall object detection performance.
