# Descriptor Matching with Convolutional Neural Networks: a Comparison to SIFT

arXiv: https://arxiv.org/abs/1405.5769 · HTML: https://mlsanity.com/paper/1405.5769/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 May 2014.

## Built on by (most cited first, 8 of 8)

- [Neural Congealing: Aligning Images to a Joint Semantic Atlas](https://mlsanity.com/paper/2302.03956/) (2023)
- [Investigating the Vision Transformer Model for Image Retrieval Tasks](https://mlsanity.com/paper/2101.03771/) (2021)
- [Pairwise Matching of Intermediate Representations for Fine-grained Explainability](https://mlsanity.com/paper/2503.22881/) (2025)
- [Registering Neural Radiance Fields as 3D Density Images](https://mlsanity.com/paper/2305.12843/) (2023)
- [Modality-Aware Feature Matching in Visual and Vision-Language Applications: A Comprehensive Survey](https://mlsanity.com/paper/2507.22791/) (2025)
- [Matching Underwater Sonar Images by the Learned Descriptor Based on Style Transfer Method](https://mlsanity.com/paper/2108.12072/) (2021)
- [Pore-scale Image Patch Dataset and A Comparative Evaluation of Pore-scale Facial Features](https://mlsanity.com/paper/2512.00381/) (2025)
- [Fine-Grained Texture Identification for Reliable Product Traceability](https://mlsanity.com/paper/2104.11548/) (2021)

## Abstract (arXiv)

Latest results indicate that features learned via convolutional neural networks outperform previous descriptors on classification tasks by a large margin. It has been shown that these networks still work well when they are applied to datasets or recognition tasks different from those they were trained on. However, descriptors like SIFT are not only used in recognition but also for many correspondence problems that rely on descriptor matching. In this paper we compare features from various layers of convolutional neural nets to standard SIFT descriptors. We consider a network that was trained on ImageNet and another one that was trained without supervision. Surprisingly, convolutional neural networks clearly outperform SIFT on descriptor matching. This paper has been merged with arXiv:1406.6909