Descriptor Matching with Convolutional Neural Networks: a Comparison to SIFT
arXiv:1405.5769
Abstract
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
This paper has been merged with arXiv:1406.6909
Cited by in corpus (22)
- FlowNet: Learning Optical Flow with Convolutional Networks
- Convolutional Neural Network-based Place Recognition
- Learning to Compare Image Patches via Convolutional Neural Networks
- Do Convnets Learn Correspondence?
- Generic 3D Representation via Pose Estimation and Matching
- Discriminative Unsupervised Feature Learning with Exemplar Convolutional Neural Networks
- FCSS: Fully Convolutional Self-Similarity for Dense Semantic Correspondence
- DXSLAM: A Robust and Efficient Visual SLAM System with Deep Features
- Domain-Size Pooling in Local Descriptors: DSP-SIFT
- Visual Scene Representations: Contrast, Scaling and Occlusion
- Unsupervised Abnormality Detection Using Heterogeneous Autonomous Systems
- What Do Deep CNNs Learn About Objects?
- Persistent Evidence of Local Image Properties in Generic ConvNets
- Matching Underwater Sonar Images by the Learned Descriptor Based on Style Transfer Method
- Image Matching via Loopy RNN
- Robust Angular Local Descriptor Learning
- Learning Transformation-Aware Embeddings for Image Forensics
- Training Deep Neural Networks to Detect Repeatable 2D Features Using Large Amounts of 3D World Capture Data
- Can We Teach Computers to Understand Art? Domain Adaptation for Enhancing Deep Networks Capacity to De-Abstract Art
- PicHunt: Social Media Image Retrieval for Improved Law Enforcement
- Fine-Grained Texture Identification for Reliable Product Traceability
- Semi-supervised learning of deep metrics for stereo reconstruction