通过正交转换层实现向后兼容的对齐表示

摘要

视觉检索系统在更新模型时面临重大挑战，因为旧表示和新表示之间存在不对齐。昂贵且资源密集的回填过程涉及在引入新模型时重新计算图库集中图像的特征向量。为解决这一问题，先前的研究探讨了向后兼容的训练方法，使得新旧表示可以直接进行比较，无需回填。尽管取得了这些进展，但在向后兼容性和独立训练模型性能之间取得平衡仍然是一个未解决的问题。本文通过扩展表示空间的附加维度并学习正交变换来解决这一问题，以实现与旧模型的兼容性，并同时整合新信息。这种变换保留了原始特征空间的几何结构，确保我们的模型与先前版本保持一致，同时学习新数据。我们的正交兼容对齐（OCA）方法在模型更新期间消除了重新索引的需要，并确保特征可以在不同模型更新之间直接进行比较，无需额外的映射函数。在CIFAR-100和ImageNet-1k上的实验结果表明，我们的方法不仅保持与先前模型的兼容性，而且实现了最先进的准确性，优于几种现有方法。

English

Visual retrieval systems face significant challenges when updating models with improved representations due to misalignment between the old and new representations. The costly and resource-intensive backfilling process involves recalculating feature vectors for images in the gallery set whenever a new model is introduced. To address this, prior research has explored backward-compatible training methods that enable direct comparisons between new and old representations without backfilling. Despite these advancements, achieving a balance between backward compatibility and the performance of independently trained models remains an open problem. In this paper, we address it by expanding the representation space with additional dimensions and learning an orthogonal transformation to achieve compatibility with old models and, at the same time, integrate new information. This transformation preserves the original feature space's geometry, ensuring that our model aligns with previous versions while also learning new data. Our Orthogonal Compatible Aligned (OCA) approach eliminates the need for re-indexing during model updates and ensures that features can be compared directly across different model updates without additional mapping functions. Experimental results on CIFAR-100 and ImageNet-1k demonstrate that our method not only maintains compatibility with previous models but also achieves state-of-the-art accuracy, outperforming several existing methods.

通过正交转换层实现向后兼容的对齐表示

Backward-Compatible Aligned Representations via an Orthogonal Transformation Layer

摘要

Summary

Support

Support