Anchor-aware Deep Metric Learning For Audio-visual Retrieval

Abstract

Metric learning minimizes the gap between similar (positive) pairs of data points and increases the separation of dissimilar (negative) pairs, aiming at capturing the underlying data structure and enhancing the performance of tasks like audio-visual cross-modal retrieval (AV-CMR). Recent works employ sampling methods to select impactful data points from the embedding space during training. However, the model training fails to fully explore the space due to the scarcity of training data points, resulting in an incomplete representation of the overall positive and negative distributions. In this paper, we propose an innovative Anchor-aware Deep Metric Learning (AADML) method to address this challenge by uncovering the underlying correlations among existing data points, which enhances the quality of the shared embedding space. Specifically, our method establishes a correlation graph-based manifold structure by considering the dependencies between each sample as the anchor and its semantic

Stats

• citations6
• S2 citations—
• github stars0
• HF likes0
• heat score6.34
• arxiv keyzeng2024anchor

Related papers

• DAS: Densely-anchored Sampling For Deep Metric Learning (2022)9.76
• Guided Deep Metric Learning (2022)6.77
• Multimodal Metric Learning For Tag-based Music Retrieval (2020)9.76
• Integrating Language Guidance Into Vision-based Deep Metric Learning (2022)14.04
• Self-supervised Auxiliary Loss For Metric Learning In Music Similarity-based Retrieval And Auto-tagging (2023)0.00
• Dynamic Sampling For Deep Metric Learning (2020)5.84
• Cross-modal Deep Metric Learning With Multi-task Regularization (2017)9.03
• Few-shot Metric Learning: Online Adaptation Of Embedding For Retrieval (2022)8.09