Abstract:Feature fusion is a commonly used strategy in image retrieval tasks, which aggregates the matching responses of multiple visual features. Feasible sets of features can be either descriptors (SIFT, HSV) for an entire image or the same descriptor for different local parts (face, body). Ideally, the to-be-fused heterogeneous features are pre-assumed to be discriminative and complementary to each other. However, the effectiveness of different features varies dramatically according to different queries. That is to say, for some queries, a feature may be neither discriminative nor complementary to existing ones, while for other queries, the feature suffices. As a result, it is important to estimate the effectiveness of features in a query-adaptive manner. To this end, this article proposes a new late fusion scheme at the score level. We base our method on the observation that the sorted score curves contain patterns that describe their effectiveness. For example, an "L"-shaped curve indicates that the feature is discriminative while a gradually descending curve suggests a bad feature. As such, this paper introduces a query-adaptive late fusion pipeline. In the hand-crafted version, it can be an unsupervised approach to tasks like particular object retrieval. In the learning version, it can also be applied to supervised tasks like person recognition and pedestrian retrieval, based on a trainable neural module. Extensive experiments are conducted on two object retrieval datasets and one person recognition dataset. We show that our method is able to highlight the good features and suppress the bad ones, is resilient to distractor features, and achieves very competitive retrieval accuracy compared with the state of the art. In an additional person re-identification dataset, the application scope and limitation of the proposed method are studied.

Node-Sensitive Graph Fusion via Topo-Correlation for Image Retrieval

Deep Hashing with Top Similarity Preserving for Image Retrieval

SIFT-Based Image Retrieval Combining the Distance Measure of Global Image and Sub-Image

Rank-Aware Graph Fusion With Contextual Dissimilarity Measurement For Image Retrieval

Adaptively Weighted Graph Fusion for Image Retrieval.

Normalized Similarity Measurement And Query Adaptive Fusion For Graph Based Visual Reranking

Query Adaptive Fusion for Graph-Based Visual Reranking.

Augmented Feature Fusion for Image Retrieval System

Robust ImageGraph: Rank-Level Feature Fusion for Image Search.

Adaptive multi-feature fusion via cross-entropy normalization for effective image retrieval

Graph-Based Features for Image Retrieval

Multi-Index Fusion Via Similarity Matrix Pooling for Image Retrieval

Social Neighborhood Graph and Multigraph Fusion Ranking for Multifeature Image Retrieval

Optimal Fusion of Multi-spectral Images under Correlation Coefficient Criteria

Multi-Feature Fusion Image Retrieval Algorithm Based on Fuzzy Color

Local Semantic Correlation Modeling Over Graph Neural Networks for Deep Feature Embedding and Image Retrieval

Query Adaptive Late Fusion for Image Retrieval.

Feature Fusion for Image Retrieval with Adaptive Bitrate Allocation and Hard Negative Mining.

Scene Graph Based Fusion Network For Image-Text Retrieval

Image Retrieval Based on Multi-feature Fusion of Color Connected Region

A new fusion approach for content based image retrieval with color histogram and local directional pattern