Combination of Multiple Global Descriptors for Image Retrieval

Abstract

Recent studies in image retrieval task have shown that ensembling different models and combining multiple global descriptors lead to performance improvement. However, training different models for the ensemble is not only difficult but also inefficient with respect to time and memory. In this paper, we propose a novel framework that exploits multiple global descriptors to get an ensemble effect while it can be trained in an end-to-end manner. The proposed framework is flexible and expandable by the global descriptor, CNN backbone, loss, and dataset. Moreover, we investigate the effectiveness of combining multiple global descriptors with quantitative and qualitative analysis. Our extensive experiments show that the combined descriptor outperforms a single global descriptor, as it can utilize different types of feature properties. In the benchmark evaluation, the proposed framework achieves the state-of-the-art performance on the CARS196, CUB200-2011, In-shop Clothes, and Stanford Online Products on image retrieval tasks. Our model implementations and pretrained models are publicly available.

References in corpus (9)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• In Defense of the Triplet Loss for Person Re-Identification
• MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
• On Calibration of Modern Neural Networks
• Hierarchical Question-Image Co-Attention for Visual Question Answering
• Heated-Up Softmax Embedding
• Large-scale Landmark Retrieval/Recognition under a Noisy and Diverse Dataset
• Attention-Aware Generalized Mean Pooling for Image Retrieval
• 2nd Place and 2nd Place Solution to Kaggle Landmark Recognition andRetrieval Competition 2019

Cited by in corpus (10)

• Semi-supervised Feature-Level Attribute Manipulation for Fashion Image Retrieval
• Team JL Solution to Google Landmark Recognition 2019
• An Effective Pipeline for a Real-world Clothes Retrieval System
• Large Scale Open-Set Deep Logo Detection
• A Benchmark on Tricks for Large-scale Image Retrieval
• Hotel Recognition via Latent Image Embedding
• PrivateMail: Supervised Manifold Learning of Deep Features With Differential Privacy for Image Retrieval
• Self-supervised asymmetric deep hashing with margin-scalable constraint
• CBIR using Pre-Trained Neural Networks
• SLADE: A Self-Training Framework For Distance Metric Learning