PCL: Proposal Cluster Learning for Weakly Supervised Object Detection
Open accessThe DOI matches its Crossref record; 65 of the 66 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
Weakly Supervised Object Detection (WSOD), using only image-level annotations to train object detectors, is of growing importance in object recognition. In this paper, we propose a novel deep network for WSOD. Unlike previous networks that transfer the object detection problem to an image classification problem using Multiple Instance Learning (MIL), our strategy generates proposal clusters to learn refined instance classifiers by an iterative process. The proposals in the same cluster are spatially adjacent and associated with the same object. This prevents the network from concentrating too much on parts of objects instead of whole objects. We first show that instances can be assigned object or background labels directly based on proposal clusters for instance classifier refinement, and then show that treating each cluster as a small new bag yields fewer ambiguities than the directly assigning label method. The iterative instance classifier refinement is implemented online using multiple streams in convolutional neural networks, where the first is an MIL network and the others are for instance classifier refinement supervised by the preceding one. Experiments are conducted on the PASCAL VOC, ImageNet detection, and MS-COCO benchmarks for WSOD. Results show that our method outperforms the previous state of the art significantly.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“Multi-instance learning is a group of supervised learning algorithms that has seen widespread application in WSOD [202-209].”
From Object Detection in 20 Years: A Survey (2023) · cites this paper as [207]
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148 · cited 4×
“Here we train a Fast R-CNN (FRCNN) [7] using the VGG16 model and the same five image scales (horizontal flips only in training).”
MethodInfluentialImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“Recently, due to the development of Convolutional Neural Network (CNN) [1], [2] and the availability of large scale datasets with detailed boundingbox-level annotations [3], [4], [5], there have been great leap forwards in object detection [6], [7], [8], [9], [10], [11].”
BackgroundFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472 · cited 1×
“Recently, due to the development of Convolutional Neural Network (CNN) [1], [2] and the availability of large scale datasets with detailed boundingbox-level annotations [3], [4], [5], there have been great leap forwards in object detection [6], [7], [8], [9], [10], [11].”
BackgroundScikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 1×
“The k-means algorithm to produce top ranking proposals is implemented by scikit-learn [56].”
MethodFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812 · cited 1×
“Recently, due to the development of Convolutional Neural Network (CNN) [1], [2] and the availability of large scale datasets with detailed boundingbox-level annotations [3], [4], [5], there have been great leap forwards in object detection [6], [7], [8], [9], [10], [11].”
BackgroundGradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 14×
“We finally visualize some success and failure detection results on VOC 2007 trainval by PCL-Ens.+FRCNN, as in Fig.”
BackgroundInfluentialYou Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 1×
“Recently, due to the development of Convolutional Neural Network (CNN) [1], [2] and the availability of large scale datasets with detailed boundingbox-level annotations [3], [4], [5], there have been great leap forwards in object detection [6], [7], [8], [9], [10], [11].”
BackgroundMicrosoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380
(2016) · Cited by 31,590
Self-citation
Cited by
Items 301–400 of 409
Learning prototypes for multiple instance learning
Sivrikaya, Ö. E., Yüksekgönül, M., & Baydoğan, M. G. (2021) · TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES · doi:10.3906/elk-2101-103 · Cited by 3
“Well known problems have also been reformulated for this particular purpose, such as common computer vision tasks like image classification [21], weakly supervised object detection [22, 23], sequence predictions [24], sentiment analysis [25] and sound event detection [26].”
BackgroundExploring A CAM - Based Approach for Weakly Supervised Fire Detection Task
Lai, L., Chen, J., Huang, H., & Wu, Q. (2021) · IEEE International Conference on e-Business Engineering (ICEBE) · doi:10.1109/icebe52470.2021.00035 · Cited by 3
Fine-grained label learning in object detection with weak supervision of captions
Wang, X., Du, Y., Verberne, S., & Verbeek, F. J. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-13592-7 · Cited by 3
“In general, this task aims to detect objects from images based on the supervision of a set of image-level labels [11, 29, 39].”
BackgroundMethodSelf-citationMa, Y., & Zhi, Z. (2022) · Advances in Multimedia · doi:10.1155/2022/1239337 · Cited by 3
“The PCL method [14] combines the advantages of MIL and DCNN models and proposes a method to learn network parameters based on candidate set clustering.”
MethodInfluentialRetractedWang, Z., Chen, Z., & Gong, C. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9897985 · Cited by 3
Caption-Aided Product Detection via Collaborative Pseudo-Label Harmonization
Dong, X. M., Zhang, G., Zhan, X., Ding, Y., Wei, Y., Lu, M., & Liang, X. (2022) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2022.3222653 · Cited by 3
“Similar to [10], [25], [35], given the image-level annotations, the detector is built based on a self-training framework that adaptively produces pseudo-labels as self-supervision during detector training.”
BackgroundMethodInfluentialZhu, M., Wan, S., Jin, P., & Zhang, P. (2022) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata55660.2022.10020414 · Cited by 3
“Specifically, the detection of our DFFNet exceeds OICR PCL [10], DCL [12], TCANet [14], MIG [15], SANet [16], and OS-DEC [19] by 12.39%, 10.70%, 8.70%, 3.07%, 3.78%, 1.79%, and 1.37%, which are significantly marginal in terms of mAP.”
MethodInfluentialWeak-shot Object Detection through Mutual Knowledge Transfer
Du, X., Wan, W., Sun, C., & Li, C. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.01884 · Cited by 3
“Further improvements are made in OICR [29] and PCL [28] meth-ods by adding instance refinement classifiers that facilitate the iterative refinement of candidate boxes.”
MethodDu, X., Diao, W., Feng, Y., Liu, C., & Sun, X. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3341061 · Cited by 4
“The problem is ameliorated in later proposed OICR [30] and PCL [31].”
BackgroundOpen-vocabulary object detection via debiased curriculum self-training
Zhang, H., Guan, D., Ke, X., Saddik, A. E., & Lu, S. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124762 · Cited by 4
Xing, P., Huang, M., Wang, C., & Cao, Y. (2024) · Electronics · doi:10.3390/electronics13234753 · Cited by 3
Negative Prototypes Guided Contrastive Learning for Weakly Supervised Object Detection
Zhang, Y., Zhu, C., Yang, G., & Chen, S. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-43415-0_3 · Cited by 3
View-consistency network for weakly supervised oriented object detection in remote sensing images
Tingting, F., Liu, B., Chen, C., & Li, X. (2024) · National Remote Sensing Bulletin · doi:10.11834/jrs.20243138 · Cited by 2
UltraWeak: Enhancing Breast Ultrasound Cancer Detection with Deformable DETR and Weak Supervision
Khan, U., Nawaz, U., & Saddik, A. E. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-73376-5_14 · Cited by 2
Wang, C., Yan, Y., Xue, J., & Wang, H. (2025) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2025.3550052 · Cited by 2
“Miao et al. [3] assume that each input image is sampled from a mixed distribution and introduce a class-balanced hierarchical refinement (CHR) method.”
InfluentialMultiple Instance Learning for Visual Grain Quality Analysis Without Instance-Level Annotation
Ezard, B., Li, L., & An, S. (2025) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw67362.2025.00532 · Cited by 3
“It has also been shown that MIL methods can be improved via online self-distillation [29, 36, 37].”
Saliency-Guided Adaptive Random Diffusion for Remote Sensing Images Restoration with Cloud and Haze
Zhang, W., Zhang, J., & Zhang, L. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755584 · Cited by 2
Structure-sensitive graph-based multiple-instance semi-supervised learning
Nunna, S. K., Bhattu, S. N., Somayajulu, D. V. L. N., & Kumar, N. V. N. (2021) · Sadhana · doi:10.1007/s12046-021-01659-4 · Cited by 2
Weakly Supervised Object Detection with Symmetry Context
Gu, X., Zhang, Q., & Lu, Z. (2022) · Symmetry · doi:10.3390/sym14091832 · Cited by 2
Lin, Z., Yang, Z., & Wang, Y. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00403 · Cited by 4
Multi-input MLP and LSTM-Based Neural Network Model for SQL Injection Detection
Sharma, V., & Kumar, S. (2023) · Algorithms for intelligent systems · doi:10.1007/978-981-19-7892-0_35 · Cited by 2
Pseudo-Labeling Enhanced by Privileged Information and Its Application to In Situ Sequencing Images
Haghighi, M., Cruz, M. C., Weisbart, E., Cimini, B. A., Singh, A., Bauman, J., Lozada, M. E., Kavari, S. L., Neal, J. T., Blainey, P. C., et al. (2023) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2023/531 · Cited by 3
“…supervised object detection (WSOD) studies the strategies to use image-level categorical annotations without location information to detect objects [Tang et al., 2018; 1 The source code and supplementary materials for this paper can be found at http://broad.io/PLePIISS Dong et al., 2021; Lin et…”
BackgroundA Quantum-Inspired Direct Learning Strategy for Positive and Unlabeled Data
Zhang, C., Du, X., & Zhang, Y. (2023) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-023-00373-9 · Cited by 2
“Here, we treat the U set as the primary source of global clustering information [8], while the P set acts as a guiding reference.”
Weakly Semi-Supervised Tool Detection in Minimally Invasive Surgery Videos
Fujii, R., Hachiuma, R., & Saitô, H. (2024) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp48485.2024.10445821 · Cited by 2
“WSOD methods [4, 5, 6] reduce the cost by replacing the box annotations with large data with cheaper weak annotations, such as image labels, which are much easier to collect than bounding box annotations.”
MethodOn-the-fly point annotation for fast medical video labeling
Meyer, A., Mazellier, J.-P., Dana, J., & Padoy, N. (2024) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-024-03098-y · Cited by 5
“WSOD methods use abundant but weakly annotated data, such as image labels [12, 21].”
No Annotations for Object Detection in Art Through Stable Diffusion
Ramos, P., Gonthier, N., Khan, S., Nakashima, Y., & García, N. (2025) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61041.2025.00607 · Cited by 6
“Baselines We compare our method against previous work on weakly-supervised object detection: • We do not re-implement the above baselines; instead, we report results as presented in previous works [16,37,55].”
Tan, J., Wang, C., Tan, X., Zhang, M., & Wang, H. (2025) · Remote Sensing · doi:10.3390/rs17101673 · Cited by 2
Activity and Relationship Modeling Driven Weakly Supervised Object Detection
Li, Y., Qian, Y., Xu, Y., & Zhang, Y. (2021) · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · doi:10.1109/icpr48806.2021.9412697 · Cited by 2
“Some other work introduced self-placed subset learning paradigm [10], iteratively proposal cluster refining [11] and bottom-up and top-down objectness distillation [12] to generate reliable supervision.”
BackgroundMethodEfficient Weakly-Supervised Object Detection With Pseudo Annotations
Yuan, Q., Sun, G., Liang, J., & Leng, B. (2021) · IEEE Access · doi:10.1109/access.2021.3099497 · Cited by 2
“We can observe that our method can obtain more tight bounding boxes and localize objects more correctly instead of only focusing on discriminative parts as our base-line method, i.e., PCL [15] does.”
BackgroundMethodInfluentialA weakly supervised object detection approach using point annotation
姚, 洁., 韩, 军., & 张, 鼎. (2021) · Scientia Sinica Informationis · doi:10.1360/ssi-2021-0089 · Cited by 2
An adaptive learning-based weakly supervised object detection via context awareness
Zeng, X., Li, Z., & Zhang, W. (2021) · International Conference on Big Data & Artificial Intelligence & Software Engineering (ICBASE) · doi:10.1109/icbase53849.2021.00068 · Cited by 2
PGTRNET: Two-Phase Weakly Supervised Object Detection with Pseudo Ground Truth Refinement
Wang, J., Zhou, H., & Yu, X. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746625 · Cited by 2
“PGTRNet adopts the PCL [16] and the Faster r-cnn [19] as the backbone networks of the first-phase and second-phase detectors, respectively.”
BackgroundMethodInfluentialSFOD-Trans: semi-supervised fine-grained object detection framework with transformer module
Liu, Q., Zhang, G., Li, K., Zhou, F., & Yu, D. (2022) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-022-02682-1 · Cited by 2
“Comparing WSCDN [44], MELM [45] and STAC [46], our method is 8.2 % higher in AP and 8.9 % higher in AR than the second place, which confirms the high competitiveness of SFOD-Trans in the detection of hepatic portal vein.”
MethodInfluentialTanaka, Y., Yoshida, S. M., & Terao, M. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9898014 · Cited by 2
“Examples include semi-supervised learning [5, 6, 7, 8] and weakly supervised learning [9, 10].”
BackgroundWeakly Supervised Point Clouds Transformer for 3D Object Detection
Tang, Z., Sun, B., Ma, T., Li, D., & Xu, Z. (2022) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc55140.2022.9921926 · Cited by 3
Cascade Attentive Dropout for Weakly Supervised Object Detection
Gao, W., Chen, Y., & Peng, Y. (2023) · Neural Processing Letters · doi:10.1007/s11063-023-11243-y · Cited by 2
“7 PCL [3] 43.”
BackgroundMethodInfluentialImage-level labeled weakly supervised object detection: a survey
Chen, Z., Wang, Z., & Gong, C. (2023) · Journal of Image and Graphics · doi:10.11834/jig.220854 · Cited by 2
Salient-aware multiple instance learning optimized network for weakly supervised object detection
Zhang, H., Wang, Y., & Yang, Y. (2024) · The Visual Computer · doi:10.1007/s00371-023-03234-w · Cited by 2
“We compare the proposed method with the eleven state-of-the-art detection models including WSDDN [1], OICR [4], PCL [5], Yang et al .”
InfluentialCheng, B., Li, Z., Li, H., Ding, Z., & Qin, T. (2021) · Remote Sensing · doi:10.3390/rs13245065 · Cited by 1
Recurrent self-optimizing proposals for weakly supervised object detection
Zhang, M., & Zeng, B. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07818-w · Cited by 1
“Most WSOD methods (e.g., [35–37, 39, 42, 44, 46]) focus on the second stage to improve the performance of weakly supervised detector, for instance, mining object instances to refine proposal scores [37, 39, 46] or combining weakly supervised detector and fully supervised detector for a joint…”
BackgroundMethodResultInfluentialDetection Of Snowball and Jersey Colour And Jersey Number Using CNN
Shubhangi, D. C., Gadgay, B., Begum, M., & Waheed, M. (2023) · International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics (IITCEE) · doi:10.1109/iitcee57236.2023.10091073 · Cited by 1
Mutual learning with discrepancy for weakly supervised object detection
Yang, K., Ye, X., Wu, Y., Li, J., Li, X., & Zhu, P. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.128453 · Cited by 1
Discriminatively Matched Part Tokens for Pointly Supervised Instance Segmentation
Guo, Z., Wan, F., Liao, M., Zhang, Y., & Ye, Q. (2025) · International Journal of Computer Vision · doi:10.1007/s11263-025-02547-8 · Cited by 1
SCIR: A Weakly Supervised Contextual Instance Refinement Method for Remote Sensing Object Detection
Yang, X., Zhou, Z., Duan, S., & Yang, D. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3640394 · Cited by 1
“Meanwhile, we keep the training settings including learning rate, mini-batch, weight decay, and momentum consistent with those in [8], [9], and [10].”
InfluentialContext-Aware Weakly Supervised Surgical Instrument Detection
Li, R., Zhao, Z., & Li, F. (2025) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-025-01787-x · Cited by 1
“…region problem, Tang et al. successively introduced online instance classifier refinement (OICR) [18] and proposal cluster learning (PCL) [19], significantly improving the performance of WSDDN by connecting optimization stages hierarchically and gradually optimizing classification scores…”
Yang, X., Zhou, Z., & Yang, D. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3647662 · Cited by 2
“Meanwhile, we keep the training settings, including learning rate, mini-batch, weight decay, and momentum, consistent with those in [6], [7], and [8].”
Dual Frequency-Domain Enhancement Based Weakly Supervised Object Detection for SAR Imagery
Nie, Y., & Duan, W. (2025) · IEEE International Conference on Unmanned Systems (ICUS) · doi:10.1109/icus66297.2025.11295018 · Cited by 1
“Contemporary SAR-agnostic WSOD research predominantly builds upon the OICR framework [10] and its enhancements, including PCL [11] which improved contextual modeling through segmentation confidence, B-OICR [12] that implemented multi-branch distillation, BUAA [13] which developed feature…”
Liu, S., & Zhang, Z. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-4445-5_13 · Cited by 1
Markaki, S., & Panagiotakis, C. (2026) · Remote Sensing · doi:10.3390/rs18030505 · Cited by 1
Weak Supervised Surface Defect Detection Method Based on Selective Search and CAM
Xiao, Y., Song, Y., & Tang, X. (2019) · Chinese Automation Congress · doi:10.1109/cac48633.2019.8996269 · Cited by 1
“Compared with PCL [17], the recall, precision and mAP increased by 5.0%, 0.1%, and 1.4% respectively, indicating that the optimization strategy proposed by our algorithm played a role.”
MethodInfluentialWeakly-Supervised Action Detection Guided by Audio Narration
Ye, K., & Kovashka, A. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00159 · Cited by 1
“Some use image-level labels [6,13,25,47,48,52,54] or captions [9,58,60] to learn object detection models.”
BackgroundMixed Supervision for Instance Learning in Object Detection with Few-shot Annotation
Zhong, Y., Wang, C., Li, S., Zhu, Z., Wang, Y., & Zheng, W. (2022) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3503161.3548242 · Cited by 3
“Tang et al. [32] and Tang et al. [33] coupled WSDDN with several instance classifiers, and proposed an online refinement method to refine several branches instance classifiers based on the output of previous branches.”
MethodA Simple Approach and Benchmark for 21,000-Category Object Detection
Lin, Y., Li, C., Cao, Y., Zhang, Z., Wang, J., Wang, L., Liu, Z., & Hu, H. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-20083-0_1 · Cited by 1
“There have been extensive studies on these topics [36,2,38,37,7,53,35,44,3,48,49,50,29,10].”
BackgroundMethodWeakly Supervised Object Localization with Noisy-Label Learning
Fan, Y., Wei, S., Tan, C., Chen, X., & Zhao, Y. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-18916-6_39 · Cited by 1
CaltechFN: Distorted and Partially Occluded Digits
Rim, P., Saha, S., & Rim, M. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-27066-6_14 · Cited by 1
Liu, Z., Wang, K., Li, C., Ding, S., & Xi, J. (2023) · IET Computer Vision · doi:10.1049/cvi2.12203 · Cited by 1
Ye, B., Zhang, J., Rao, Y., Gao, R., & Zeng, D. (2024) · International Journal of Remote Sensing · doi:10.1080/01431161.2024.2359734 · Cited by 1
“…2015), and Faster R-CNN (Ren et al. 2015), as well as weakly supervised models like WSDDN (Bilen and Vedaldi 2016), OICR (Tang et al. 2017), PCL (Tang et al. 2020), MELM (Wan et al. 2018), DCL (Yao et al. 2020), PCIR (Feng et al. 2020a), MIG (B. Wang, Zhao, and Li 2021), CDN (Huo et al. 2023),…”
MethodInfluentialLocal Context-Aware for Weakly Supervised Object Detection
Chen, P., Nie, X., Li, T., Lu, X., & Bi, Y. (2024) · International Conference on Cyber-Physical Social Intelligence (ICCSI) · doi:10.1109/iccsi62669.2024.10799353 · Cited by 1
2D-Based Coarse-to-Fine Approaches for Small Target Segmentation in Abdominal CT Scans
Zhou, Y., Yu, Q., Wang, Y., Xie, L., Shen, W., Fishman, E. K., & Yuille, A. (2019) · Advances in computer vision and pattern recognition · doi:10.1007/978-3-030-13969-8_3 · Cited by 1
Self-citationLearning Maximally Predictive Prototypes in Multiple Instance LearningPreprint
Yüksekgönül, M., Sivrikaya, Ö. E., & Baydoğan, M. G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1910.00965 · Cited by 1
“…for this particular purpose, such as common computer vision tasks like image classification [Wu et al., 2015], weakly supervised object detection [Tang et al., 2018, Wan et al., 2019]; sequence predictions [Dennis et al., 2018], sentiment analysis [Angelidis and Lapata, 2017] and sound event…”
BackgroundTwo-phase weakly supervised object detection with pseudo ground truth miningPreprint
Wang, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.00231 · Cited by 1
“This model improves the mAP by 4% compared with the single PCL [15] model.”
BackgroundMethodInfluentialWeakly supervised object detection based on deep metric learning
Dong, W., Zhang, X., & Zhao, P. (2021) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1748/2/022015
“[8] proposed a multiple streams detector optimization method for the WSOD model.”
MethodCategory‐wise feature extractor based on ADL method for weak‐supervised object localisation
Hu, Y., Zhu, D., Ai, X., & Xu, Y. (2020) · IET Image Processing · doi:10.1049/iet-ipr.2020.0640
Deep Learning based Face Mask Recognition System -A Review
Hari Priya, K., & Malathi, S. (2021) · International Conference on Inventive Computation Technologies (ICICT) · doi:10.1109/icict50816.2021.9358515
Learning to recognize objects by adaptive knowledge transfer
Tao, Q. (2021) · DR-NTU (Nanyang Technological University) · doi:10.32657/10356/151507
PGTRNet: Two-phase Weakly Supervised Object Detection with Pseudo Ground Truth Refining.Preprint
Wang, J., Zhou, H., & Yu, X. (2021) · arXiv (Cornell University)
Self-Supervised Object Detection via Generative Image SynthesisPreprint
Mustikovela, S. K., Mello, S. D., Prakash, A., Iqbal, U. H., Liu, S., Nguyen-Phuoc, T., Rother, C., & Kautz, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.09848
“Compared to Wetectron (mAP of 38.1 for All) and PCL (mAP of 33.2 for All), SSOD (mAP of 68.4 for All) has ∼2X better detection accuracy.”
MethodInfluentialDeep heterogeneous superpixel neural networks for image analysis and feature extraction
Yang, Z. (2021) · MOspace Institutional Repository (University of Missouri) · doi:10.32469/10355/88061
Border Sensitive Network in Weakly Supervised Thyroid Nodule Detection for Ultrasound Image
Luo, T., Xu, T., Yu, J. X., Li, X., Wei, X., Yu, M., Zhang, R., Gao, J., & Yu, R. (2021) · IEEE International Conference on Bioinformatics and Biomedicine (BIBM) · doi:10.1109/bibm52615.2021.9669500
Strong Representation Learning for Weakly Supervised Object Detection
Yu, L. M. S., Weidong, D., He, Y., Gou, Y., Wu, Z., & Lv, Y. (2022) · International Journal Of Engineering And Computer Science · doi:10.18535/ijecs/v11i02.4650
A Self-Training Method with a Noise-Robust Loss Function for Weakly Supervised Object DetectionPreprint
Wan, Y., Li, X., Gao, Y., Gao, L., & Wen, L. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4201034
Weakly Supervised Object Detection with Position Information of Convolution Neural Network
Sun, B., Yan, H., He, J. Y. D., Yu, L., & Zhang, Y. (2022) · IEEE International Conference on Information, Communication and Networks (ICICN) · doi:10.1109/icicn56848.2022.10006528
“Modifying the loss function [10]-[13] or designing a more reasonable training process can optimize the solving process and help model find the global optimal solution.”
BackgroundMethodWeakly supervised object detection with interactive edge attentive collaboration
Gao, W., Chen, Y., & Peng, Y. (2022) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc55256.2022.10033877
“0 In addition, in order to prove the extensibility of our method, we embed the edge attentive collaboration module into PCL* network[24] to obtain EAC-NET*, and conduct related experiments on the PASCAL VOC 2007 and VOC 2012 datasets again.”
MethodWenlong, 高. G., Ying, 陈. C., & Yong, 彭. P. (2023) · Laser & Optoelectronics Progress · doi:10.3788/lop212868
Wang, Y., & Huo, H. (2023) · Academic Journal of Science and Technology · doi:10.54097/ajst.v5i3.7806
Jiang, W., & Ou, L. (2023) · International Conference on Computer Supported Cooperative Work in Design (CSCWD) · doi:10.1109/cscwd57460.2023.10152646
“On the basis of OICR, in order to further improve the detection performance, Tang et al. [13] propose a method for clustering the deep features of the candidate regions so that the candidate regions with similar features are clustered together as much as possible.”
BackgroundMethodAutomated Labelling of Traffic Police Data Based on Weak Supervision
Zheng, Y., Zhang, H., Du, J., Hu, C., & Bao, H. (2023) · International Conference on Dependable Systems and Their Applications (DSA) · doi:10.1109/dsa59317.2023.00093
“To address this issue, Peng Tang et al. [9] introduced a clustering algorithm to group highly overlapping regions into a new set for training the classifier.”
BackgroundP2W: Improving the Cognition of Object Wholeness for Weakly Supervised Object Detection
Fang, L., Dang, J., & Tang, X. (2023) · China Automation Congress (CAC) · doi:10.1109/cac59555.2023.10451736
A Study of Object Detection Based on Weakly Supervised Learning
Wang, G. (2024) · International Journal of Computer Science and Information Technology · doi:10.62051/ijcsit.v2n1.50
“[8] mainly used domain migration technology to convert the source domain data in the source domain into images with example labels, or to generate corresponding pseudo-label labels for the target domain data through pseudo-labeling.”
MethodQin, Y., Wang, Z., Qin, Y., & Shao, C. (2024) · Lecture notes in electrical engineering · doi:10.1007/978-981-97-3682-9_80
Local-to-Global Point Supervised Object Detector via Aggregation of Discriminative Parts
Zhang, Y., Hou, Y., Liu, X., Li, X., Mou, F., Zhang, P., Qi, X., Jia, J., Wang, L., & Zhao, X. (2024) · Lecture notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering · doi:10.1007/978-3-031-71716-1_12
Sketch2Seg: Sketch-Based Image Segmentation with Pre-trained Diffusion Model
Dai, X., Deng, H., Li, K., & Qi, Y. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-78347-0_3
Instance-Dependent Noise Refinement in Segment Anything Model for Weakly Supervised Object Detection
Taherkhani, F., & Kazemi, E. (2024) · Lecture notes in computer science · doi:10.1007/978-981-96-0885-0_20 · Cited by 1
“For comparison with other methods as per [86,85], we use data augmentation, including multi-level scaling and horizontal flipping during training.”
Pointflow: Point Supervised Video Object DetectionPreprint
Wang, K., Yu, X., Han, X., Li, G., Huang, Z., & Han, Z. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5342130
Zhu, Y., Wang, S., & Sun, Z. (2025) · Lecture notes in computer science · doi:10.1007/978-981-96-6594-5_23
Weakly Supervised Object Detection Framework based on Classification-Localization Consistency
Zhu, Y., Wang, S., Lu, M., & Sun, Z. (2025) · IEEE International Conference on Multimedia and Expo (ICME) · doi:10.1109/icme59968.2025.11209030
“Output: Pseudo box supervision P label ( c, l ) In Stage 2, we employed WSDDN [10] and PCL [9] as weakly supervised detectors, respectively.”
Learning to Detect Sleep Micro-Events From Coarse Sleep Stage Annotations
Wang, C., Pei, Y., Hu, J., Han, C., Xu, J., Zhang, L., Yu, F. N., Jin, B., & Luo, W. (2025) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2025.3633639
“A key challenge of weakly supervised learning with coarse sleep-stagelabelsiseventfragmentation[39],inwhichaground-truthmicro-eventisbrokenintoseveralshort,disjointsegments becausethenetworkattendsonlytolocalwaveformsnippets.”
Guo, Y., He, H., Chen, Z., Huang, Z., Lu, R., Shi, L., Wang, Z., & Zhang, W. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3710491
“We also weight the cross-entropy loss for each proposal like prior works [32], [37].”
InfluentialPlug-and-Play Global and Local Collaborative Fusion for Weakly Supervised Object DetectionPreprint
Liang, Q., & Zhang, Y. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5615210
Plug-and-Play Global and Local Collaborative Fusion for Weakly Supervised Object DetectionPreprint
Liang, Q., & Zhang, Y. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5959469
Weakly Supervised Rotated Object Detection Based on Multi-Scale Attention Mechanism
Fan, S., Han, Y., Zhou, W., Peng, J., Jia, M., Fu, Y., & Zhou, J. (2025) · IEEE Intl Conf on Dependable, Autonomic and Secure Computing, Intl Conf on Pervasive Intelligence and Computing, Intl Conf on Cloud and Big Data Computing, Intl Conf on Cyber Science and Technology Congress (DASC/PiCom/CBDCom/CyberSciTech) · doi:10.1109/cyberscitech68397.2025.00066
Cross-Image Co-Attentive Instance Mining for Weakly Supervised Object Detection
Li, D. (2025) · IEEE International Conference on Industrial Electronics for Sustainable Energy Systems (IESES) · doi:10.1109/ieses66335.2025.11359908
Roy, I. S., Acharjya, K., & Singh, A. K. (2025) · International Conference on IoT, Communication and Automation Technology (ICICAT) · doi:10.1109/icicat68430.2025.11414712
“Semi-supervised and weakly supervised learning reduce dependence on annotations, achieving human-like performance [17].”
Practical weakly supervised object detection: benchmark and method
Cao, H., & Chen, Z. (2026) · Neural Computing and Applications · doi:10.1007/s00521-026-11842-5
Plug-and-Play global and local collaborative fusion for weakly supervised object detection
Liang, Q., & Zhang, Y. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.115857
Annotation-efficient learning for object discovery and detection
Vo, H. V. (2022) · theses.fr (ABES) · doi:10.70675/aae8cd95z8e92z4d35z9384zafe56efdad87
Query-guided feature mining for weakly supervised object detection
Fang, Y., Xu, X., Li, W., Wang, Y., Jia, H., Shen, Y., Xie, J., & Lin, S. (2026) · Neurocomputing · doi:10.1016/j.neucom.2026.133698
Geometry-driven perception and action for topology-agnostic robotic rebar tying
Wang, H., Jin, Z., & Shao, Y. (2026) · Automation in Construction · doi:10.1016/j.autcon.2026.107013
Towards Point-Supervised Open World Object DetectionPreprint
Jia, S., Bai, S., Ma, R., Wu, T., Zeng, T., Zhang, G., & Li, Z. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.6852393
Self-citationSemi-supervised YOLO-DEP for high-resolution X-ray component localization and counting
Xiao, Z., Sun, H., Xu, T. L., & Li, L. (2026) · Journal of X-Ray Science and Technology · doi:10.1177/08953996261439069
“To mitigate annotation cost, weakly-supervised counting 54 – 56 and semi-supervised detection 57 – 60 have been explored in natural scenes.”
Metadata
- Authors
- Tang, P., Wang, X., Bai, S., Shen, W., Bai, X., Liu, W., & Yuille, A.
- Affiliations
- Huazhong University of Science and Technology (CN)
- University of Oxford (GB)
- Shanghai University (CN)
- Johns Hopkins University (US)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2020-01-01
- Volume
- 42
- Issue
- 1
- Pages
- 176-191
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- 1939-3539
- OpenAlex
- W2813911573
- Cited by
- 406
- References
- 62
- Citations by year
- 2026: 16 · 2025: 41 · 2024: 54 · 2023: 59 · 2022: 74 · 2021: 82 · 2020: 50
- FWCI
- 20.27
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (61733007, 61572207, 61876212, 61672336)
- ONR (N00014-15-1-2356)
- Hubei Scientific and Technical Innovation Key Project
- HUST Academic Frontier Youth Team
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- weakly supervised object detection, multiple instance learning, image-level annotations, iterative refinement, multi-stream convolutional neural network
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.