Bayesian Triplet Loss: Uncertainty Quantification In Image Retrieval

Abstract

Uncertainty quantification in image retrieval is crucial for downstream decisions, yet it remains a challenging and largely unexplored problem. Current methods for estimating uncertainties are poorly calibrated, computationally expensive, or based on heuristics. We present a new method that views image embeddings as stochastic features rather than deterministic features. Our two main contributions are (1) a likelihood that matches the triplet constraint and that evaluates the probability of an anchor being closer to a positive than a negative; and (2) a prior over the feature space that justifies the conventional l2 normalization. To ensure computational efficiency, we derive a variational approximation of the posterior, called the Bayesian triplet loss, that produces state-of-the-art uncertainty estimates and matches the predictive performance of current state-of-the-art methods.

Stats

• citations33
• S2 citations—
• github stars0
• HF likes0
• heat score11.49
• arxiv keywarburg2020bayesian

Related papers

• Unsupervised Data Uncertainty Learning In Visual Retrieval Systems (2019)0.00
• Bayesian Metric Learning For Uncertainty Quantification In Image Retrieval (2023)0.00
• Ranking-aware Uncertainty For Text-guided Image Retrieval (2023)0.00
• Exploring Uncertainty In Conditional Multi-modal Retrieval Systems (2019)0.00
• Uncertainty-based Cross-modal Retrieval With Probabilistic Representations (2022)0.00
• Exploring Uncertainty Measures For Image-caption Embedding-and-retrieval Task (2019)2.26
• Risk Controlled Image Retrieval (2023)0.00
• Reliability-aware Prediction Via Uncertainty Learning For Person Image Retrieval (2022)8.35