mlsanityData to 14 Oct 2026

Streamlining the Development of Active Learning Methods in Real-World Object Detection

arXiv 2508.19906 ↗First posted 27 Aug 2025cs.CVcited by 3

Not on the map. Why?

Abstract (arXiv)

Active learning (AL) for real-world object detection faces computational and reliability challenges that limit practical deployment. Developing new AL methods requires training multiple detectors across iterations to compare against existing approaches. This creates high costs for autonomous driving datasets where the training of one detector requires up to 282 GPU hours. Additionally, AL method rankings vary substantially across validation sets, compromising reliability in safety-critical transportation systems. We introduce object-based set similarity (OSS), a metric that addresses these challenges. OSS (1) quantifies AL method effectiveness without requiring detector training by measuring similarity between training sets and target domains using object-level features. This enables the elimination of ineffective AL methods before training. Furthermore, OSS (2) enables the selection of representative validation sets for robust evaluation. We validate our similarity-based approach on three autonomous driving datasets (KITTI, BDD100K, CODA) using uncertainty-based AL methods as a case study with two detector architectures (EfficientDet, YOLOv3). This work is the first to unify AL training and evaluation strategies in object detection based on object similarity. OSS is detector-agnostic, requires only labeled object crops, and integrates with existing AL pipelines. This provides a practical framework for deploying AL in real-world applications where computational efficiency and evaluation reliability are critical. Code is available at https://mos-ks.github.io/publications/.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 30 of its 56 references that are in the map. First posted on arXiv 27 Aug 2025.

Builds on

30 of 56 references.

References · 30

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2018YOLOv3: An Incremental Improvementcited by 2,659
  4. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,352
  5. 2018BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learningcited by 1,288
  6. 2019EfficientDet: Scalable and Efficient Object Detectioncited by 1,020
  7. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 913
  8. 2020A Survey of Deep Active Learningcited by 554
  9. 2019Learning Loss for Active Learningcited by 336
  10. 2014Efficient Object Localization Using Convolutional Networkscited by 227
  11. 2022Active Learning on a Budget: Opposite Strategies Suit High and Low Budgetscited by 133
  12. 2019Consistency-based Semi-supervised Active Learning: Towards Minimizing Labeling Costcited by 114
18 more
  1. 2022CODA: A Real-World Road Corner Case Dataset for Object Detection in Autonomous Drivingcited by 100
  2. 2022A Deeper Look into Aleatoric and Epistemic Uncertainty Disentanglementcited by 87
  3. 2021Active Learning for Deep Object Detection via Probabilistic Modelingcited by 79
  4. 2021Multiple instance active learning for object detectioncited by 77
  5. 2018Practical Obstacles to Deploying Active Learningcited by 68
  6. 2020Scalable Active Learning for Object Detectioncited by 58
  7. 2022Entropy-based Active Learning for Object Detection with Progressive Diversity Constraintcited by 53
  8. 2018Active Learning for Deep Object Detectioncited by 50
  9. 2018Localization-Aware Active Learning for Object Detectioncited by 48
  10. 2019Parting with Illusions about Deep Active Learningcited by 38
  11. 2021Consistency-based Active Learning for Object Detectioncited by 32
  12. 2021TALISMAN: Targeted Active Learning for Object Detection with Rare Classes and Slices using Submodular Mutual Informationcited by 27
  13. 2018Deep Ensemble Bayesian Active Learning : Addressing the Mode Collapse issue in Monte Carlo dropout via Ensemblescited by 20
  14. 2020On Initial Pools for Deep Active Learningcited by 18
  15. 2022ALBench: A Framework for Evaluating Active Learning in Object Detectioncited by 7
  16. 2023Overcoming the Limitations of Localization Uncertainty: Efficient & Exact Non-Linear Post-Processing and Calibrationcited by 5
  17. 2022Towards Rapid Prototyping and Comparability in Active Learning for Deep Object Detectioncited by 3
  18. 2022Complementing Semi-Supervised Learning with Uncertainty Quantificationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026From Model Uncertainty to Human Attention: Localization-Aware Visual Cues for Scalable Annotation Reviewcited by 0
  2. 2026Distance to Class Prototypes: Active Learning for Object Detectioncited by 0
  3. 2026Generative Verification: Rethinking the Uncertainty Signal for Active Learning of Object Detectioncited by 0