Learning-Based Dimensionality Reduction for Computing Compact and Effective Local Feature Descriptors

A distinctive representation of image patches in form of features is a key component of many computer vision and robotics tasks, such as image matching, image retrieval, and visual localization. State-of-the-art descriptors, from hand-crafted descriptors such as SIFT to learned ones such as HardNet, are usually high-dimensional; 128 dimensions or even more. The higher the dimensionality, the larger the memory consumption and computational time for approaches using such descriptors. In this paper, we investigate multi-layer perceptrons (MLPs) to extract low-dimensional but high-quality descriptors. We thoroughly analyze our method in unsuper-vised, self-supervised, and supervised settings, and evaluate the dimensionality reduction results on four representative descriptors. We consider different applications, including visual localization, patch verification, image matching and retrieval. The experiments show that our lightweight MLPs trained using supervised method achieve better dimensionality reduction than PCA. The lower-dimensional descriptors generated by our approach outperform the original higher-dimensional descriptors in downstream tasks, especially for the hand-crafted ones. The code is available at https://github.com/PRBonn/descriptor-dr.

Authors

• Hao Dong 0011
• Xieyuanli Chen
• Mihai Dusmanu
• Viktor Larsson
• Marc Pollefeys
• Cyrill Stachniss

Keywords

• Dimensionality reduction
• Location awareness
• Visualization
• Runtime
• Image matching
• Memory management
• Supervised learning
• Local Descriptors
• Local Feature Descriptors
• Image Registration
• Multilayer Perceptron
• Visual Images
• Image Patches
• Image Retrieval
• Memory Consumption
• Visual Localization
• Visual Matching
• Hidden Layer
• Learning Strategies
• Unsupervised Methods
• Latent Space
• Cluster Assignment
• Evaluation Protocol
• Dimensionality Reduction Methods
• Classification Layer
• Robotic Applications
• Descriptor Space
• Distance Loss
• Reconstruction Loss
• Triplet Loss
• Original Ones
• Low-dimensional Embedding
• Viewpoint Changes
• Matching Task
• Illumination Changes
• Linear Projection