Abstract
The shape of objects is an important source of visual information in a wide range of applications. One of the core challenges of shape quantification is to ensure that the extracted measurements remain invariant to transformations that preserve an object's intrinsic geometry, such as changing its size, orientation, and position in the image. In this work, we introduce ShapeEmbed, a self-supervised representation learning framework designed to encode the contour of objects in 2D images, represented as a Euclidean distance matrix, into a shape descriptor that is invariant to translation, scaling, rotation, reflection, and point indexing. Our approach overcomes the limitations of traditional shape descriptors while improving upon existing state-of-the-art autoencoder-based approaches. We demonstrate that the descriptors learned by our framework outperform their competitors in shape classification tasks on natural and biological images. We envision our approach to be of particular relevance to biological imaging applications.
Abstract (translated)
物体的形状是许多应用中重要的视觉信息来源。量化形状的一个核心挑战在于确保提取出的测量值在保持对象固有几何不变的情况下（如改变其大小、方向和图像中的位置）仍然具有不变性。在这项工作中，我们介绍了ShapeEmbed，这是一个自我监督表示学习框架，旨在将二维图像中物体轮廓（以欧几里得距离矩阵的形式表示）编码为一种形状描述符，这种描述符在平移、缩放、旋转、反射和平点索引下保持不变。我们的方法克服了传统形状描述符的局限性，并且改进了现有的基于自编码器的最佳实践。我们展示了通过我们框架学习到的描述符在自然图像和生物图像中的形状分类任务上优于其竞争对手的表现。我们认为，我们的方法特别适用于生物成像应用。
URL
https://arxiv.org/abs/2507.01009
