Anchor Box Optimization for Object Detection
Open accessThe DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index.
Abstract
In this paper, we propose a general approach to optimize anchor boxes for object detection. Nowadays, anchor boxes are widely adopted in state-of-the-art detection frameworks. However, these frameworks usually pre-define anchor box shapes in heuristic ways and fix the sizes during training. To improve the accuracy and reduce the effort of designing anchor boxes, we propose to dynamically learn the anchor shapes, which allows the anchors to automatically adapt to the data distribution and the network learning capability. The learning approach can be easily implemented with stochastic gradient descent and can be plugged into any anchor box-based detection framework. The extra training cost is almost negligible and it has no impact on the inference time or memory cost. Exhaustive experiments demonstrate that the proposed anchor optimization method consistently achieves significant improvement (≥ 1% mAP absolute gain) over the baseline methods on several benchmark datasets including Pascal VOC 07+12, MS COCO and Brainwash. Meanwhile, the robustness is also verified towards different anchor initialization methods and the number of anchor shapes, which greatly simplifies the problem of anchor box design.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“6 faster rcnn res[5] 600 76.”
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 · cited 10×
“To handle the background class, one can use an extra background class in the cross entropy loss, e.g. in [13, 18].”
BackgroundMethodInfluentialYou 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 5×
“The YOLO [15] belongs to this spectrum but was improved by YOLOv2 [16] with anchor-based approach.”
BackgroundMethodInfluentialMicrosoft 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
(2016) · Cited by 31,590
Feature Pyramid Networks for Object Detection
Lin, T.-Y., Dollar, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.106 · Cited by 30,939 · cited 1×
“Extensions to other detectors should be straightforward, such as the RPN of Faster R-CNN [18], SSD [13], Feature Pyramid Network (FPN) [9], and RetinaNet [11].”
MethodFocal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.324 · Cited by 30,581 · cited 6×
“Extensions to other detectors should be straightforward, such as the RPN of Faster R-CNN [18], SSD [13], Feature Pyramid Network (FPN) [9], and RetinaNet [11].”
BackgroundMethodInfluentialBatch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Ioffe, S., & Szegedy, C. (2015) · arXiv · doi:10.48550/arxiv.1502.03167 · Cited by 27,818 · cited 1×
“To further reduce the learning diffi-culties, we apply the batch normalization [5] on the output of ∆ wi and ∆ hi without the shifting parameters.”
MethodYOLO9000: Better, Faster, Stronger
Redmon, J., & Farhadi, A. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.690 · Cited by 19,892 · cited 10×
“To calculate the variables δ i,j , we normally use the IoU as the metric [18, 13, 16].”
BackgroundMethodInfluentialThe pascal visual object classes (VOC) challenge
Everingham, M., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2010) · International Journal of Computer Vision · doi:10.1007/s11263-009-0275-4 · Cited by 19,599 · cited 2×
“The PASCAL VOC dataset [3] contains box annotations over 20 object categories.”
Method
Cited by
RepPoints: Point Set Representation for Object Detection
Yang, Z., Liu, S., Hu, H., Wang, L., & Lin, S. (2019) · 2019 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2019.00975 · Cited by 1,307
“The bestperforming object detectors to date generally follow a multistage recognition paradigm [24, 4, 13, 23, 2, 35, 27], and the bounding box representation appears in almost all stages: 1) as pre-defined [33, 25] or learnt [40, 44, 46] anchors that serve as hypotheses over the bounding box space; 2) as refined obje…”
BackgroundAli, M. L., & Zhang, Z. (2024) · Computers · doi:10.3390/computers13120336 · Cited by 383
“The architecture of RFB Net is shown in Figure 5.”
Automated Defect Detection From Ultrasonic Images Using Deep Learning
Medak, D., Posilović, L., Subašić, M., Budimir, M., & Lončarić, S. (2021) · IEEE Transactions on Ultrasonics Ferroelectrics and Frequency Control · doi:10.1109/tuffc.2021.3081750 · Cited by 127
“There are several approaches[30], [31], [32] that can be used to estimate good anchors hyperparameters but it is often needed to make some assumptions about the objects’ shapes so that the calculation would be possible.”
Mammogram breast cancer CAD systems for mass detection and classification: a review
Hassan, N. M., Hamad, S. H., & Mahar, K. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-12332-1 · Cited by 118
“Anchor boxes are considered as the key concept behind both of those techniques, it’s one of the main factors that affect the performance of the detector in detecting the objects within the image [90].”
BackgroundA novel finetuned YOLOv6 transfer learning model for real-time object detection
Gupta, C., Gill, N. S., Gulia, P., & Chatterjee, J. M. (2023) · Journal of Real-Time Image Processing · doi:10.1007/s11554-023-01299-3 · Cited by 72
“Classification of the object is done using the anchor box[27].”
Method
Metadata
- Authors
- Zhong, Y., Wang, J., Peng, J., & Zhang, L.
- Affiliations
- University of Illinois Urbana-Champaign (US)
- Microsoft (United States) (US)
- Venue
- 2020 IEEE Winter Conference on Applications of Computer Vision (WACV)
- Publisher
- IEEE
- Published
- 2020-03
- Pages
- 1275-1283
- ISSN
- 2472-6737 (ISSN-L)
- OpenAlex
- W3010357916
- Cited by
- 118
- References
- 27
- Citations by year
- 2026: 7 · 2025: 19 · 2024: 23 · 2023: 15 · 2022: 17 · 2021: 20 · 2020: 12
- FWCI
- 4.94
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Winter Conference on Applications of Computer Vision (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- anchor box optimization, object detection, Pascal VOC, MS COCO
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Anchor Box Optimization for Object Detection. https://app.cochat.ai/papers/10.1109/wacv45572.2020.9093498
- The paper
- Zhong, Y., Wang, J., Peng, J., & Zhang, L. (2020). Anchor Box Optimization for Object Detection. 2020 IEEE Winter Conference on Applications of Computer Vision (WACV). https://doi.org/10.1109/wacv45572.2020.9093498
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.