Lune

CVPR2024Top-tier venue

Learning for Transductive Threshold Calibration in Open-World Recognition

Qin Zhang, Dongsheng An, Tianjun Xiao, Tong He, Qingming Tang, Ying Nian Wu, Joseph Tighe, Yifan Xing

Abstract

In deep metric learning for visual recognition, the calibration of distance thresholds is crucial for achieving desired model performance in the true positive rates (TPR) or true negative rates (TNR). However, calibrating this threshold presents challenges in open-world scenarios, where the test classes can be entirely disjoint from those encountered during training. We define the problem of finding distance thresholds for a trained embedding model to achieve target performance metrics over unseen open-world test classes as open-world threshold calibration. Existing posthoc threshold calibration methods, reliant on inductive inference and requiring a calibration dataset with a similar distance distribution as the test data, often prove ineffective in open-world scenarios. To address this, we introduce OpenGCN, a Graph Neural Network-based transductive threshold calibration method with enhanced adaptability and robustness. OpenGCN learns to predict pairwise connectivity for the unlabeled test instances embedded in a graph to determine its TPR and TNR at various distance thresholds, allowing for transductive inference of the distance thresholds which also incorporates test-time information. Extensive experiments across open-world visual recognition benchmarks validate OpenGCN's superiority over existing posthoc calibration methods for open-world threshold calibration.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Adaptive Conformal Inference Under Distribution Shift cites Classification with Valid and Adaptive Coverage
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on15

  1. Calibrating Deep Neural Networks using Focal LossJishnu Mukhoti, Viveka Kulharia, Amartya Sanyal, Stuart Golodetz et al.NeurIPS 2020 · 674 citations
  2. Adaptive Conformal Inference Under Distribution ShiftIsaac Gibbs, Emmanuel J. CandèsNeurIPS 2021 · 665 citations
  3. Classification with Valid and Adaptive CoverageYaniv Romano, Matteo Sesia, Emmanuel J. CandèsNeurIPS 2020 · 586 citations
  4. Open-World Semi-Supervised LearningKaidi Cao, Maria Brbic, Jure LeskovecICLR 2022 · 246 citations
  5. Transductive Episodic-Wise Adaptive Metric for Few-Shot LearningLimeng Qiao, Yemin Shi, Jia Li, Yonghong Tian et al.ICCV 2019 · 196 citations

Related papers

  1. EGonc : Energy-based Open-Set Node Classification with substitute UnknownsQin Zhang, Zelin Shi, Shirui Pan, Junyang Chen et al.NeurIPS 2024 · 2 citations
  2. Be Confident! Towards Trustworthy Graph Neural Networks via Confidence CalibrationXiao Wang, Hongrui Liu, Chuan Shi, Cheng YangNeurIPS 2021 · 158 citations
  3. EVINET: Towards Open-World Graph Learning via Evidential Reasoning NetworkWeijie Guan, Haohui Wang, Jian Kang, Lihui Liu et al.KDD 2025
  4. negMIX: Negative Mixup for OOD Generalization in Open-Set Node ClassificationJunwei Gong, Xiao Shen, Zhihao Chen, Shirui Pan et al.WWW 2026
  5. Graph Out-of-Distribution Generalization via Causal InterventionQitian Wu, Fan Nie, Chenxiao Yang, Tianyi Bao et al.WWW 2024 · 58 citations
Dusk over the sea between two cliffs drawn in fine vertical lines