Domain-Size Pooling in Local Descriptors: DSP-SIFT

Abstract

We introduce a simple modification of local image descriptors, such as SIFT, based on pooling gradient orientations across different domain sizes, in addition to spatial locations. The resulting descriptor, which we call DSP-SIFT, outperforms other methods in wide-baseline matching benchmarks, including those based on convolutional neural networks, despite having the same dimension of SIFT and requiring no training.

arXiv commentExtended version of the CVPR 2015 paper. Technical Report UCLA CSD 140022

References in corpus (8)

• Return of the Devil in the Details: Delving Deep into Convolutional Nets
• Descriptor Matching with Convolutional Neural Networks: a Comparison to SIFT
• Dense Correspondences Across Scenes and Scales
• Unsupervised feature learning by augmenting single images
• Steps Towards a Theory of Visual Information: Active Perception, Signal-to-Symbol Conversion and the Interplay Between Sensing and Control
• Multi-scale Orderless Pooling of Deep Convolutional Activation Features
• Visual Scene Representations: Contrast, Scaling and Occlusion
• On the Design and Analysis of Multiple View Descriptors

Cited by in corpus (3)

• Visual Representations: Defining Properties and Deep Approximations
• See the Difference: Direct Pre-Image Reconstruction and Pose Estimation by Differentiating HOG
• Discriminate-and-Rectify Encoders: Learning from Image Transformation Sets