Large-Scale Image Retrieval with Attentive Deep Local Features
Preprint
Abstract
We propose an attentive local feature descriptor suitable for large-scale image retrieval, referred to as DELF (DEep Local Feature). The new feature is based on convolutional neural networks, which are trained only with image-level annotations on a landmark image dataset. To identify semantically useful local features for image retrieval, we also propose an attention mechanism for keypoint selection, which shares most network layers with the descriptor. This framework can be used for image retrieval as a drop-in replacement for other keypoint detectors and descriptors, enabling more accurate feature matching and geometric verification. Our system produces reliable confidence scores to reject false positives effectively---in particular, it is robust against queries that have no correct match in the database. To evaluate the proposed descriptor, we introduce a new large-scale dataset, which involves challenges in both database and query such as background clutter, partial occlusion, multiple landmarks, objects in variable scales, etc. We show that DELF outperforms the state-of-the-art global and local descriptors in the large-scale setting by significant margins.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples2016
- Stacked Attention Networks for Image Question Answering2016
- Learning Deep Features for Discriminative Localization2016
- MatchNet: Unifying feature and metric learning for patch-based matching2015
- Product Quantization for Nearest Neighbor SearchIEEE Transactions on Pattern Analysis and Machine Intelligence, 2010
- Speeded-Up Robust Features (SURF)Computer Vision and Image Understanding, 2008
- Lost in quantization: Improving particular object retrieval in large scale image databases2008
- Hamming Embedding and Weak Geometric Consistency for Large Scale Image Search2008
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004
- Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography1987