CoChat
Sign Up
Cited by 155Machine Learning and AlgorithmsAdvanced Neural Network ApplicationsDomain Adaptation and Few-Shot Learning

Multiple Instance Active Learning for Object Detection

Open access
Yuan, Tianning, Wan, Fang, Fu, Mengying, Liu, Jianzhuang, Xu, Songcen, Ji, Xiangyang, Ye, Qixiang·2021·2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
DOI 10.1109/cvpr46437.2021.00529
Open this paper in CoChat — free accountOpen PDF

Abstract

Despite the substantial progress of active learning for image recognition, there still lacks an instance-level active learning method specified for object detection. In this paper, we propose Multiple Instance Active Object Detection (MI-AOD), to select the most informative images for detector training by observing instance-level uncertainty. MI-AOD defines an instance uncertainty learning module, which leverages the discrepancy of two adversarial instance classifiers trained on the labeled set to predict instance uncertainty of the unlabeled set. MI-AOD treats unlabeled images as instance bags and feature anchors in images as instances, and estimates the image uncertainty by re-weighting instances in a multiple instance learning (MIL) fashion. Iterative instance uncertainty learning and re-weighting facilitate suppressing noisy instances, toward bridging the gap between instance uncertainty and image-level uncertainty. Experiments validate that MI-AOD sets a solid baseline for instance-level active learning. On commonly used object detection datasets, MI-AOD outperforms state-of-the-art methods with significant margins, particularly when the labeled sets are small. Code is available at https://github.com/yuantn/MI-AOD.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr46437.2021.00529

Cites

4 of 4 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Compared to state-of-the-art methods, our approach yields up to 7.7% relative mAP improvement for PASCAL-VOC and up to 7% for MS-COCO.

    “Compared to state-of-the-art methods [4-8], our approach yields up to a 7 . 7% and 7% relative mAP improvement for PASCAL-VOC and MS-COCO, respectively.”

    From Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detection (2021) · cites this paper as [8]

  • In the first active learning cycle on VOC07+12, our method has a relative improvement over the random baseline by 10.5%, and over 8.2% compared to the best overall active learning method.

    “We observe that in the first active learning cycle, our method has a relative improvement over the random baseline by 10 . 5% , and over 8 . 2% compared to the best overall active learning method [8].”

    From Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detection (2021) · cites this paper as [8]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 40

  1. Oisin Mac Aodha, Neill D. F. Campbell, Jan Kautz, and Gabriel J. Brostow. Hierarchical subquery evaluation for ac- tive learning on a graph. In CVPR, pages 564-571, 2014.
  2. Hakan Bilen and Andrea Vedaldi. Weakly supervised deep detection networks. In CVPR, pages 2846-2854, 2016. 5
  3. Marc-Andre Carbonneau, Eric Granger, and Ghyslain Gagnon. Bag-level aggregation for multiple-instance active learning in instance classification problems. IEEE TNNLS, 30(5):1441-1451, 2019. 2
  4. Ehsan Elhamifar, Guillermo Sapiro, Allen Y. Yang, and S. Shankar Sastry. A convex optimization framework for ac- tive learning. In ICCV, pages 209-216, 2013. 2
  5. Alexander Freytag, Erik Rodner, and Joachim Denzler. Se- lecting influential examples: Active learning with expected model output changes. In ECCV, pages 562-577, 2014. 2
  6. Denis A. Gudovskiy, Alec Hodgkinson, Takuya Yamaguchi, and Sotaro Tsukizawa. Deep active learning for biased datasets via fisher kernel self-supervision. In CVPR, pages 9038-9046, 2020. 3
  7. Yuhong Guo. Active instance sampling via matrix partition. In NeurIPS, pages 802-810, 2010. 2
  8. Mahmudul Hasan and Amit K. Roy-Chowdhury. Context aware active learning of activity recognition models. In ICCV, pages 4543-4551, 2015. 2
  9. Sheng-Jun Huang, Nengneng Gao, and Songcan Chen. Multi-instance multi-label active learning. In IJCAI, pages 1886-1892, 2017. 2
  10. Ajay J. Joshi, Fatih Porikli, and Nikolaos Papanikolopou- los. Multi-class active learning for image classification. In CVPR, pages 2372-2379, 2009. 2
  11. Christoph Käding, Erik Rodner, Alexander Freytag, and Joachim Denzler. Active and continuous exploration with deep neural networks and expected model output changes. arXiv preprint arXiv:1612.06129, 2016. 2 · arXiv
  12. David D. Lewis and Jason Catlett. Heterogeneous uncer- tainty sampling for supervised learning. In Machine Learn- ing, pages 148-156, 1994. 2
  13. David D. Lewis and William A. Gale. A sequential algorithm for training text classifiers. In SIGIR, pages 3-12, 1994. 2
  14. Liang Lin, Keze Wang, Deyu Meng, Wangmeng Zuo, and Lei Zhang. Active self-paced learning for cost-effective and progressive face identification. IEEE TPAMI, 40(1):7-19, 2018. 2
  15. Tsung-Yi Lin, Priya Goyal, Ross B. Girshick, Kaiming He, and Piotr Dollár. Focal loss for dense object detection. IEEE TPAMI, 42(2):318-327, 2020. 3, 4, 6
  16. Binghao Liu, Yao Ding, Jianbin Jiao, Ji Xiangyang, and Qix- iang Ye. Anti-aliasing semantic reconstruction for few-shot semantic segmentation. In CVPR, 2021. 1
  17. Binghao Liu, Jianbin Jiao, and Qixiang Ye. Harmonic feature activation for few-shot semantic segmentation. IEEE TIP, pages 3142-3153, 2021. 1
  18. Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, and Alexander C Berg. Ssd: Single shot multibox detector. In ECCV, pages 21-37, 2016. 6
  19. Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott E. Reed, Cheng-Yang Fu, and Alexander C. Berg. SSD: single shot multibox detector. In ECCV, pages 21-37, 2016. 6
  20. Zhao-Yang Liu and Sheng-Jun Huang. Active sampling for open-set classification without initial annotation. In AAAI, pages 4416-4423, 2019. 2
  21. Zhao-Yang Liu, Shao-Yuan Li, Songcan Chen, Yao Hu, and Sheng-Jun Huang. Uncertainty aware graph gaussian process for semi-supervised learning. In AAAI, pages 4957-4964, 2020. 2
  22. Wenjie Luo, Alexander G. Schwing, and Raquel Urtasun. Latent structured active learning. In NeurIPS, pages 728- 736, 2013. 2
  23. Hieu Tat Nguyen and Arnold W. M. Smeulders. Active learn- ing using pre-clustering. In ICML, 2004. 2
  24. Dan Roth and Kevin Small. Margin-based active learning for structured output spaces. In ECML, pages 413-424, 2006. 2
  25. Nicholas Roy and Andrew McCallum. Toward optimal ac- tive learning through sampling estimation of error reduction. In ICML, pages 441-448, 2001. 2
  26. Ozan Sener and Silvio Savarese. Active learning for con- volutional neural networks: A core-set approach. In ICLR, 2018. 2, 6
  27. Burr Settles and Mark Craven. An analysis of active learn- ing strategies for sequence labeling tasks. In EMNLP, pages 1070-1079, 2008. 2
  28. Burr Settles, Mark Craven, and Soumya Ray. Multiple- instance active learning. In NeurIPS, pages 1289-1296, 2007. 2
  29. Samarth Sinha, Sayna Ebrahimi, and Trevor Darrell. Vari- ational adversarial active learning. In ICCV, pages 5971- 5980, 2019. 2
  30. Andrews Stuart, Tsochantaridis Ioannis, and Hofmann Thomas. Support vector machines for multiple-instance learning. In NeurIPS, pages 561-568, 2002. 5
  31. Ying-Peng Tang and Sheng-Jun Huang. Self-paced active learning: Query the right thing at the right time. In AAAI, pages 5117-5124, 2019. 2
  32. Fang Wan, Chang Liu, Wei Ke, Xiangyang Ji, Jianbin Jiao, and Qixiang Ye. C-mil: Continuation multiple instance learning for weakly supervised object detection. In CVPR, pages 2199-2208, 2019. 2
  33. Fang Wan, Pengxu Wei, Zhenjun Han, Jianbin Jiao, and Qix- iang Ye. Min-entropy latent model for weakly supervised object detection. IEEE TPAMI, 41(10):2395-2409, 2019. 2
  34. Keze Wang, Dongyu Zhang, Ya Li, Ruimao Zhang, and Liang Lin. Cost-effective active learning for deep image classification. IEEE TCSVT, 27(12):2591-2600, 2017. 2
  35. Ran Wang, Xi-Zhao Wang, Sam Kwong, and Chen Xu. Incorporating diversity and informativeness in multiple- instance active learning. IEEE TFS, 25(6):1460-1475, 2017.
  36. Yi Yang, Zhigang Ma, Feiping Nie, Xiaojun Chang, and Alexander G. Hauptmann. Multi-class active learning by uncertainty sampling with diversity maximization. IJCV, 113(2):113-127, 2015. 2
  37. Donggeun Yoo and In So Kweon. Learning loss for active learning. In CVPR, pages 93-102, 2019. 2, 3, 6
  38. Beichen Zhang, Liang Li, Shijie Yang, Shuhui Wang, Zheng- Jun Zha, and Qingming Huang. State-relabeling adversarial active learning. In CVPR, pages 8753-8762, 2020. 2
  39. Xiaosong Zhang, Fang Wan, Chang Liu, Rongrong Ji, and Qixiang Ye. Freeanchor: Learning to match anchors for vi- sual object detection. In NeurIPS, pages 147-155, 2019. 3
  40. Xiaosong Zhang, Fang Wan, Chang Liu, Xiangyang Ji, and Qixiang Ye. Learning to match anchors for visual object de- tection. IEEE TPAMI, pages 1-14, 2021.

Cited by

Sorted by citations

Authors
Yuan, T., Wan, F., Fu, M., Liu, J., Xu, S., Ji, X., & Ye, Q.
Affiliations
  • University of Chinese Academy of Sciences (CN)
  • Huawei Technologies (China) (CN)
  • Tsinghua University (CN)
ORCID
Venue
2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2021-06
Pages
5326-5335
ISSN
1063-6919 (ISSN-L)
DOI
10.1109/cvpr46437.2021.00529
OpenAlex
W3184557372
Cited by
155
References
4
Citations by year
2026: 18 · 2025: 30 · 2024: 37 · 2023: 43 · 2022: 24 · 2021: 3
FWCI
12.36
Type
conference paper
Language
en
Open access
Licence
Locations
  • IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Funders
Chinese Academy of Sciences
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Algorithms
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
Keywords
active learning (machine learning), object detection, multiple instance learning

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review