Weakly Supervised Object Detection Using Proposal- and Semantic-Level Relationships
The DOI matches its Crossref record; 56 of the 58 references it cites were found in a registry or an index.
Abstract
In recent years, weakly supervised object detection has attracted great attention in the computer vision community. Although numerous deep learning-based approaches have been proposed in the past few years, such an ill-posed problem is still challenging and the learning performance is still behind the expectation. In fact, most of the existing approaches only consider the visual appearance of each proposal region but ignore to make use of the helpful context information. To this end, this paper introduces two levels of context into the weakly supervised learning framework. The first one is the proposal-level context, i.e., the relationship of the spatially adjacent proposals. The second one is the semantic-level context, i.e., the relationship of the co-occurring object categories. Therefore, the proposed weakly supervised learning framework contains not only the cognition process on the visual appearance but also the reasoning process on the proposal- and semantic-level relationships, which leads to the novel deep multiple instance reasoning framework. Specifically, built upon a conventional CNN-based network architecture, the proposed framework is equipped with two additional graph convolutional network-based reasoning models to implement object location reasoning and multi-label reasoning within an end-to-end network training procedure. Comprehensive experiments on the widely used PASCAL VOC and MS COCO benchmarks have been implemented, which demonstrate the superior capacity of the proposed approach when compared with other state-of-the-art methods and baseline models.
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 [206]
Citations
Cites
ImageNet 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
Faster 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
Semi-supervised classification with graph convolutional networks
Kipf, T. N., & Welling, M. (2016) · arXiv · doi:10.48550/arxiv.1609.02907 · Cited by 24,853
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.74 · Cited by 24,513
Selective Search for Object Recognition
Uijlings, J. R. R., van de Sande, K. E. A., Gevers, T., & Smeulders, A. W. M. (2013) · International Journal of Computer Vision · doi:10.1007/s11263-013-0620-5 · Cited by 5,980
Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
Yan, S., Xiong, Y., & Lin, D. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.12328 · Cited by 5,410
DeepGCNs: Can GCNs Go As Deep As CNNs?
Li, G., Muller, M., Thabet, A., & Ghanem, B. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00936 · Cited by 1,400
Multi-Label Image Recognition With Graph Convolutional Networks
Chen, Z.-M., Wei, X.-S., Wang, P., & Guo, Y. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00532 · Cited by 1,261
Exploring Visual Relationship for Image Captioning
Yao, T., Pan, Y., Li, Y., & Mei, T. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01264-9_42 · Cited by 870
Videos as Space-Time Region Graphs
Wang, X., & Gupta, A. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01228-1_25 · Cited by 765
Research works printed but not linked: 4
- 10.1109/CVPR.2016.311 · DOI · Paper page
- 10.1109/ICCV.2019.00662 · DOI
- Wei, TS2C: Tight box mining with surrounding segmentation context for weakly supervised object detection, 434
- 10.1109/CVPR.2019.00964 · DOI
Cited by
Items 101–132 of 132
Understanding Image Retrieval Re-Ranking: A Graph Neural Network Perspective
Zhang, X., Jiang, M., Zheng, Z., Tan, X., & Yang, Y. (2026) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3803010 · Cited by 67
Neighbor patches merging reduces spatial redundancy to accelerate vision transformer
Jiang, K., Peng, P., Lian, Y., Shao, W., & Xu, W. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128733 · Cited by 3
Online Progressive Instance-Balanced Sampling for Weakly Supervised Vibration Damper Detection
Chen, M., Tian, Y., Li, Z., Li, E., & Liang, Z. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3273655 · Cited by 5
Lai, Q., Vong, C., Shi, S.-Q., & Chen, C. L. P. (2024) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2024.3436853 · Cited by 5
“Following the previous literature [20, 32, 34], the experiments are implemented on the VOC 2007 dataset.”
Weakly- and Semisupervised Probabilistic Segmentation and Quantification of Reverberation Artifacts
Hung, A. L. Y., Chen, E., & Galeotti, J. (2022) · BME Frontiers · doi:10.34133/2022/9837076 · Cited by 6
An Integrated Framework for Pavement Crack Segmentation and Severity Estimation
Alsharayah, O., Manasreh, D., & Nazzal, M. D. (2026) · Buildings · doi:10.3390/buildings16030677 · Cited by 2
Object Affinity Learning: Towards Annotation-Free Instance Segmentation
Wang, Y., Chen, Y., & Zhang, Z. (2023) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2023.3298351 · Cited by 3
Adversarially Robust Object Detection via Deviation Calibration and Content Preservation
Zhang, X., Peng, B., Lei, J., Xue, C. G., Yao, Y., & Huang, Q. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3526460 · Cited by 2
“At present, DNN-based object detection methods have been studied and promising progress has been made [28], [29], [30].”
A 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
Dual-aware Domain Mining and Cross-aware Supervision for Weakly-supervised Semantic Segmentation
Guo, Y., Liang, X., Wu, B., Zheng, X., & Zhang, X. (2023) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3589343 · Cited by 2
Image-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
“To find more complete bounding boxes, some methods were used, such as context information [7, 26, 28, 35], refinement [4, 5], bounding box regression [8, 9], multitasking [19, 23], and boost learning [30, 31].”
SDLNet: Statistical Deep Learning Network for Co-Occurring Object Detection and Identification
Singh, B. K., & Lobo, N. D. V. (2024) · International Conference on Machine Learning Technologies (ICMLT) · doi:10.1145/3674029.3674055 · Cited by 1
Truncated and integrated class activation maps for weakly supervised defect detection
Dong, H., Liu, B., Dong, Y., & Liu, G. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/adcf3d · Cited by 2
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
Robust AUC maximization for classification with pairwise confidence comparisons
Shi, H., Xie, M., & Huang, S.-J. (2023) · Frontiers of Computer Science · doi:10.1007/s11704-023-2709-5 · Cited by 1
Local 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
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
Learning to Segment Blob-like Objects by Image-Level Counting
Wüstefeld, K., Ebbinghaus, R., & Weichert, F. (2023) · Applied Sciences · doi:10.3390/app132212219
DOSDet: A Sequential Feature Enhancement Network for Deck Oil Spill Detection
Zhang, H., Chen, F., An, H., Ma, Y., Zheng, Z., Yue, S., Liu, B., Tong, J., Wang, Y., & Hou, D. (2024) · International Conference on Electronics and Communication, Network and Computer Technology (ECNCT) · doi:10.1109/ecnct63103.2024.10704525
“Zhang D. et al. [2] [3] proposed two levels of context into weakly supervised learning frameworks, resulting in a new deep multiple-instance reasoning framework.”
BackgroundDual-Environment Dehazing Using Enhanced U-Net for Indoor and Outdoor Scenarios
Bhowmik, S., Sharma, A., & Dangi, R. (2025) · IEEE International Students' Conference on Electrical, Electronics and Computer Science (SCEECS) · doi:10.1109/sceecs64059.2025.10940251
“The primary goal of this technique is to restore clean image from its hazy counterpart, thus reducing the negative effects of air haze on several visual tasks such as scene comprehension[1], [2] object recognition[3], image classification [4], [5] and image segmentation.”
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
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.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
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
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
A surrogate ensemble fusion framework for robust watermark extraction in black-box models
Zhang, Z., Luo, S., Xing, F., & Pan, L. (2026) · Information Fusion · doi:10.1016/j.inffus.2026.104547 · Cited by 1
Through the Looking Glass: A Dual Perspective on Weakly Supervised Few-Shot Segmentation
Ma, J., Xie, G.-S., Zhao, F., & Li, Z. (2026) · IEEE Transactions on Image Processing · doi:10.1109/tip.2026.3708050
“Weakly-supervised few-shot semantic segmentation (WFSS) has garnered increasing attention in recent years due to its alignment with the complex and diverse characteristics of real-world application scenarios [40], [41], [42], [43], [44], [45], [46].”
Sparse weakly semi-supervised learning for oriented ship detection in polarimetric SAR images
He, Y., Gao, G., Zhang, T., & He, D. (2026) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2026.08.025
Factor Graph-Optimized Cross-Scale Detector for Wafer Defect Detection
Chen, R., Long, B., Dai, Z., Zhan, W., & Li, R. (2026) · IEEE International Test Conference in Asia (ITC-Asia) · doi:10.1109/itc-asia71864.2026.11710940
“Nevertheless, prevalent detection architectures exhibit inherent limitations [7].”
Metadata
- Authors
- Zhang, D., Zeng, W., Yao, J., & Han, J.
- Affiliations
- Northwestern Polytechnical University (CN)
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2022-06-01
- Volume
- 44
- Issue
- 6
- Pages
- 3349-3363
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- 1939-3539
- OpenAlex
- W3117854388
- Cited by
- 132
- References
- 58
- Citations by year
- 2026: 9 · 2025: 14 · 2024: 20 · 2023: 51 · 2022: 33
- FWCI
- 8.65
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- Funders
- Key R&D Program of Guangdong Province (2019B010110001)
- National Natural Science Foundation of China (61876140, U1801265)
- Research Funds for Interdisciplinary subject
- 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, graph convolutional networks
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.