What Makes for Effective Detection Proposals?
Open accessThe DOI matches its Crossref record; 69 of the 69 references it cites were found in a registry or an index.
Abstract
Current top performing object detectors employ detection proposals to guide the search for objects, thereby avoiding exhaustive sliding window search across images. Despite the popularity and widespread use of detection proposals, it is unclear which trade-offs are made when using them during object detection. We provide an in-depth analysis of twelve proposal methods along with four baselines regarding proposal repeatability, ground truth annotation recall on PASCAL, ImageNet, and MS COCO, and their impact on DPM, R-CNN, and Fast R-CNN detection performance. Our analysis shows that for object detection improving proposal localisation accuracy is as important as improving recall. We introduce a novel metric, the average recall (AR), which rewards both high recall and good localisation and correlates surprisingly well with detection performance. Our findings show common strengths and weaknesses of existing methods, and provide insights and metrics for selecting and tuning proposal methods.
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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“It is noteworthy that the Recall-to-IoU metric is just loosely [19], [20], [21] related to the ultimate detection accuracy.”
From Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2017) · cites this paper as [20]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In the case of sliding window based detectors, the imbalance between backgrounds and objects could be as extreme as 10 7 :1 [71].”
From Object Detection in 20 Years: A Survey (2023) · cites this paper as [71]
Experimental — accuracy not yet measured
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
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 1×
“Indeed, there has been some very recent work in this space [60], [61], [62] that shows promising”
BackgroundFaster 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×
“Indeed, there has been some very recent work in this space [60], [61], [62] that shows promising”
BackgroundImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Microsoft 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
Self-citationRich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.81 · Cited by 32,684
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105
Robust Real-Time Face Detection
Viola, P., & Jones, M. J. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000013087.49260.fb · Cited by 14,097
Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2015.2389824 · Cited by 11,971
Object Detection with Discriminatively Trained Part-Based Models
Felzenszwalb, P. F., Girshick, R. B., McAllester, D., & Ramanan, D. (2010) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.167 · Cited by 9,853
The Pascal Visual Object Classes Challenge: A Retrospective
Everingham, M., Eslami, S. M. A., Van Gool, L., Williams, C. K. I., Winn, J., & Zisserman, A. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-014-0733-5 · Cited by 7,139 · cited 3×
“We test the generalisation of proposal methods by evaluating on datasets with many additional classes in §4.”
Method
Research works printed but not linked: 7
- [28] A. Humayun, F. Li, and J. M. Rehg, 'Rigor: Recycling inference in graph cuts for generating object regions,' in CVPR , 2014. -
- [22] P. Krähenbühl and V. Koltun, 'Geodesic object proposals,' in ECCV , 2014. -
- [51] P. Dollár, R. Appel, and W. Kienzle, 'Crosstalk cascades for framerate pedestrian detection,' in ECCV , 2012. -
- [59] K. Tang, A. Joulin, L.-J. Li, and L. Fei-Fei, 'Co-localization in realworld images,' in CVPR , 2014. -
- [57] S. Vicente, C. Rother, and V. Kolmogorov, 'Object cosegmentation,' in CVPR , 2011. -
- [54] N. Chavali, H. Agrawal, A. Mahendru, and D. Batra, 'Objectproposal evaluation protocol is 'gameable',' arXiv:1505.05836 , 2015. - · arXiv
- szegedy, 0, Scalable, high-quality object detection, arXiv 1412 1441
Cited by
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
“In the following, we train Fast R-CNN using 2k RPN proposals, but evaluate different numbers of proposals at test-time.”
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
“In the following, we train Fast R-CNN using 2k RPN proposals, but evaluate different numbers of proposals at test-time.”
MethodHe, K., Gkioxari, G., Dollar, P., & Girshick, R. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.322 · Cited by 29,137
“Related Work R-CNN: The Region-based CNN (R-CNN) approach [10] to bounding-box object detection is to attend to a manageable number of candidate object regions [33, 16] and evaluate convolutional networks [20, 19] independently on each RoI.”
Girshick, R. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.169 · Cited by 28,105
The Cityscapes Dataset for Semantic Urban Scene Understanding
Cordts, M., Omran, M., Ramos, S., Rehfeld, T., Enzweiler, M., Benenson, R., Franke, U., Roth, S., & Schiele, B. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.350 · Cited by 12,433
“Many of the commonly used bounding box proposal methods [28, 54] first generate a set of overlapping segments, e.”
MethodSelf-citationFocal Loss for Dense Object Detection
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollar, P. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2858826 · Cited by 9,556
“contain all objects while filtering out the majority of negative locations [28], and the second stage classifies the proposals into foreground classes/background.”
BackgroundObject Detection in 20 Years: A Survey
Zou, Z., Chen, K., Shi, Z., Guo, Y., & Ye, J. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2023.3238524 · Cited by 3,689
He, K., Gkioxari, G., Dollar, P., & Girshick, R. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2844175 · Cited by 3,457
“The Region-based CNN (R-CNN) approach [8] to bounding-box object detection is to attend to a manageable number of candidate object regions [9], [10] and evaluate convolutional networks [11], [12] independently on each RoI.”
MethodMulti-view 3D Object Detection Network for Autonomous Driving
Chen, X., Ma, H., Wan, J., Li, B., & Xia, T. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.691 · Cited by 3,450
PointRCNN: 3D Object Proposal Generation and Detection From Point Cloud
Shi, S., Wang, X., & Li, H. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00086 · Cited by 3,093
“Although the recall of proposals are loosely [11, 8] related to the final 3D object detection performance, RoIs # Recall(IoU=0.”
BackgroundDeep Learning for Generic Object Detection: A Survey
Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2018) · arXiv · doi:10.48550/arxiv.1809.02165 · Cited by 2,965
Deep Learning for Generic Object Detection: A Survey
Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-019-01247-4 · Cited by 2,965
Region-based convolutional networks for accurate object detection and segmentation
Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2016) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2015.2437384 · Cited by 2,912
“Finally, we give a head-to-head comparison of R-CNNs with the recently proposed OverFeat [19] detection system.”
BackgroundInfluentialGirshick, R. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1504.08083 · Cited by 276
Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art
Zhang, L., Zhang, L., & Du, B. (2016) · IEEE Geoscience and Remote Sensing Magazine · doi:10.1109/mgrs.2016.2540798 · Cited by 2,208
“The baseline technique is the sliding window method [100], which slides the bounding box over the whole image with a small stride to generate a number of proposals.”
MethodKirillov, A., He, K., Girshick, R., Rother, C., & Dollár, P. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00963 · Cited by 1,659
A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection
Cai, Z., Fan, Q., Feris, R. S., & Vasconcelos, N. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_22 · Cited by 1,605
Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks
Bell, S., Zitnick, C. L., Bala, K., & Girshick, R. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.314 · Cited by 1,292
“Using dynamic pooling [13] we can efficiently evaluate thousands of different candidate ROIs with a single forwards pass of the network.”
MethodAn End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical Features
He, Y., Song, K., Meng, Q., & Yan, Y. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2019.2915404 · Cited by 1,341
Monocular 3D Object Detection for Autonomous Driving
Chen, X., Kundu, K., Zhang, Z., Ma, H., Fidler, S., & Urtasun, R. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.236 · Cited by 1,019
“Metrics: We evaluate our class-dependent proposals using best achievable (oracle) recall following [22, 45].”
MethodLarge scale deep learning for computer aided detection of mammographic lesions
Kooi, T., Litjens, G., van Ginneken, B., Gubern-Mérida, A., Sánchez, C. I., Mann, R., den Heeten, A., & Karssemeijer, N. (2017) · Medical Image Analysis · doi:10.1016/j.media.2016.07.007 · Cited by 965
HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection
Kong, T., Yao, A., Chen, Y., & Sun, F. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.98 · Cited by 934
“The detection accuracy with respect to mAP is shown in Table 5.”
MethodSimple Does It: Weakly Supervised Instance and Semantic Segmentation
Khoreva, A., Benenson, R., Hosang, J., Hein, M., & Schiele, B. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.181 · Cited by 764
“The intuition is that the convnet training might benefit from trading-off lower recall (more ignore pixels) for higher precision (more pixels are correctly labelled).”
BackgroundSelf-citationLearning to Refine Object Segments
Pinheiro, P. O., Lin, T.-Y., Collobert, R., & Dollár, P. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46448-0_5 · Cited by 731
1Salient Object Detection: A Survey
Borji, A., Jiang, H., & Li, J. (2016) · Cited by 761
A Comparative Analysis of Object Detection Metrics with a Companion Open-Source Toolkit
Padilla, R., Passos, W. L., Dias, T. L. B., Netto, S. L., & da Silva, E. A. B. (2021) · Electronics · doi:10.3390/electronics10030279 · Cited by 735
“The AR is defined as twice the area under the Rc IOU ( o ) × o curve for the IOU interval [ 0.5, 1 ] , that is, The authors in [90] also give a straightforward equation for the computation of the above integral from the discrete sample set, as twice the average of the excess IOU for all the…”
MethodInfluentialLearning to Segment Object Candidates
Pinheiro, P. O., Collobert, R., & Dollár, P. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1506.06204 · Cited by 648
“These methods achieve top results on object detection (when coupled with R-CNNs [10]) and also obtain the best AR [13].”
BackgroundMethodInfluentialSalient object detection: A survey
Borji, A., Cheng, M.-M., Hou, Q., Jiang, H., & Li, J. (2019) · Computational Visual Media · doi:10.1007/s41095-019-0149-9 · Cited by 524
MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving
Teichmann, M., Weber, M., Zollner, M., Cipolla, R., & Urtasun, R. (2018) · 2018 IEEE Intelligent Vehicles Symposium (IV) · doi:10.1109/ivs.2018.8500504 · Cited by 681
(2015) · Cited by 656
Gupta, A., Anpalagan, A., Guan, L., & Khwaja, A. S. (2021) · Array · doi:10.1016/j.array.2021.100057 · Cited by 565
Multiscale Combinatorial Grouping for Image Segmentation and Object Proposal Generation
Pont-Tuset, J., Arbelaez, P., T.Barron, J., Marques, F., & Malik, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2537320 · Cited by 581
“[46] provided further evidence to the interest of MCG among the current state-of-the-art in object proposal generation.”
BackgroundSurvey on semantic segmentation using deep learning techniques
Lateef, F., & Ruichek, Y. (2019) · Neurocomputing · doi:10.1016/j.neucom.2019.02.003 · Cited by 559
“Proposals are generated by algorithms ([59] provide an in-depth analysis) that are semantically meaningful and related to objects.”
BackgroundDrone-Based Object Counting by Spatially Regularized Regional Proposal Network
Hsieh, M.-R., Lin, Y.-L., & Hsu, W. H. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.446 · Cited by 547
“As the metric of recall at IoU of 0.5 is not predictive of detection accuracy, proposals with high recall but at low overlap are not effective for detection [12].”
BackgroundInfluential3D Object Detection for Autonomous Driving: A Survey
Qian, R., Lai, X., & Li, X. (2022) · Pattern Recognition · doi:10.1016/j.patcog.2022.108796 · Cited by 454
TURN TAP: Temporal Unit Regression Network for Temporal Action Proposals
Gao, J., Yang, Z., Sun, C., Chen, K., & Nevatia, R. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.392 · Cited by 498
Understanding of Object Detection Based on CNN Family and YOLO
Du, J. (2018) · Journal of Physics: Conference Series · doi:10.1088/1742-6596/1004/1/012029 · Cited by 477
“Then, with the help of warping proposal behind the R-CNN’s last convolutional layer, Spatial Pyramid Pooling Network (SPPNet) [17] accelerated its test speed by 10 times and training speed by 3 times than those of R-CNN.”
MethodDeep Extreme Cut: From Extreme Points to Object Segmentation
Maninis, K.-K., Caelles, S., Pont-Tuset, J., & Van Gool, L. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00071 · Cited by 450
Liu, Z., Wang, H., Weng, L., & Yang, Y. (2016) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2016.2565705 · Cited by 439
“5 to 1, which rewards both high recall and good localization and correlates surprisingly well with the performance of the followed detector [15].”
BackgroundSST: Single-Stream Temporal Action Proposals
Buch, S., Escorcia, V., Shen, C., Ghanem, B., & Niebles, J. C. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.675 · Cited by 433
Instance-Sensitive Fully Convolutional Networks
Dai, J., He, K., Li, Y., Ren, S., & Sun, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46466-4_32 · Cited by 413
3D Object Proposals Using Stereo Imagery for Accurate Object Class Detection
Chen, X., Kundu, K., Zhu, Y., Ma, H., Fidler, S., & Urtasun, R. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2706685 · Cited by 421
“We also report average recall (AR) [32], which has been shown to be highly correlated with the object detection”
MethodObject Contour Detection with a Fully Convolutional Encoder-Decoder Network
Yang, J., Price, B., Cohen, S., Lee, H., & Yang, M.-H. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.28 · Cited by 383
“It is established in [23, 42] to benchmark the quality of bounding box and segmented object proposals.”
BackgroundDAPs: Deep Action Proposals for Action Understanding
Escorcia, V., Caba Heilbron, F., Niebles, J. C., & Ghanem, B. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46487-9_47 · Cited by 400
“These proposal algorithms retrieve high-quality candidate regions that are likely to contain an object (high recall), before classification is performed [8, 29, 15].”
BackgroundMethodInfluentialLiang, X. (2019) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.12425 · Cited by 398
Deep learning-based object detection in low-altitude UAV datasets: A survey
Mittal, P., Singh, R., & Sharma, A. (2020) · Image and Vision Computing · doi:10.1016/j.imavis.2020.104046 · Cited by 342
Learning to Track for Spatio-Temporal Action Localization
Weinzaepfel, P., Harchaoui, Z., & Schmid, C. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.362 · Cited by 278
“We use EdgeBoxes [43], as they obtain a high recall even when considering relatively few proposals [13].”
MethodEar density estimation from high resolution RGB imagery using deep learning technique
Madec, S., Jin, X., Lu, H., De Solan, B., Liu, S., Duyme, F., Heritier, E., & Baret, F. (2019) · Agricultural and Forest Meteorology · doi:10.1016/j.agrformet.2018.10.013 · Cited by 312
EuroCity Persons: A Novel Benchmark for Person Detection in Traffic Scenes
Braun, M., Krebs, S., Flohr, F., & Gavrila, D. M. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2897684 · Cited by 307
“detection performance is shown in [48] for a lot of different proposal methods like selective search [49], MCG [50] and BING [51].”
MethodFully Convolutional Region Proposal Networks for Multispectral Person Detection
Konig, D., Adam, M., Jarvers, C., Layher, G., Neumann, H., & Teutsch, M. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2017.36 · Cited by 287
“In order to accelerate the exhaustive search, proposal generators [18] were introduced that apply fast screening with sliding windows using simple features and objectness measures that are prone to produce many false positives.”
Automatic handgun detection alarm in videos using deep learning
Olmos, R., Tabik, S., & Herrera, F. (2018) · Neurocomputing · doi:10.1016/j.neucom.2017.05.012 · Cited by 270
Weapon Detection in Real-Time CCTV Videos Using Deep Learning
Bhatti, M. T., Khan, M. G., Aslam, M., & Fiaz, M. J. (2021) · IEEE Access · doi:10.1109/access.2021.3059170 · Cited by 256
“Instead of considering all possible regions of the input frame as possibilities, this method uses detection proposal techniques to select regions [57].”
MethodFast Temporal Activity Proposals for Efficient Detection of Human Actions in Untrimmed Videos
Heilbron, F. C., Niebles, J. C., & Ghanem, B. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.211 · Cited by 276
“As such, the average recall measure summarizes proposal performance across tIoU thresholds and it correlates with detection performance as shown in [12].”
BackgroundMethodVisual place recognition: A survey from deep learning perspective
Zhang, X., Wang, L., & Su, Y. (2021) · Pattern Recognition · doi:10.1016/j.patcog.2020.107760 · Cited by 261
Hamaguchi, R., Fujita, A., Nemoto, K., Imaizumi, T., & Hikosaka, S. (2018) · 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2018.00162 · Cited by 242
A MultiPath Network for Object Detection
Zagoruyko, S., Lerer, A., Lin, T.-Y., Pinheiro, P., Gross, S., Chintala, S., & Dollar, P. (2016) · Procedings of the British Machine Vision Conference 2016 · doi:10.5244/c.30.15 · Cited by 228
Meta Faster R-CNN: Towards Accurate Few-Shot Object Detection with Attentive Feature Alignment
Han, G., Huang, S., Ma, J., He, Y., & Chang, S.-F. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i1.19959 · Cited by 211
Parametric Exponential Linear Unit for Deep Convolutional Neural Networks
Trottier, L., Giguere, P., & Chaib-draa, B. (2017) · 2017 16th IEEE International Conference on Machine Learning and Applications (ICMLA) · doi:10.1109/icmla.2017.00038 · Cited by 200
Convolutional Oriented Boundaries: From Image Segmentation to High-Level Tasks
Maninis, K.-K., Pont-Tuset, J., Arbelaez, P., & Van Gool, L. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2700300 · Cited by 231
Aziz, L., Haji Salam, M. S. B., Sheikh, U. U., & Ayub, S. (2020) · IEEE Access · doi:10.1109/access.2020.3021508 · Cited by 204
“However, this undoubtedly requires improvement in the accuracy of localization, recall, speed, and repeatability for future detection proposals [241].”
BackgroundSemantic Instance Segmentation via Deep Metric Learning
Fathi, A., Wojna, Z., Rathod, V., Wang, P., Song, H. O., Guadarrama, S., & Murphy, K. P. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.10277 · Cited by 182
“The second is to plot the recall vs the IoU threshold for a fixed number of proposals; the area under this curve is known as the “average recall” or AR [12].”
BackgroundJiao, L., Dong, S., Zhang, S., Xie, C., & Wang, H. (2020) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2020.105522 · Cited by 201
Zhu, Y., Tian, Y., Metaxas, D., & Dollar, P. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.320 · Cited by 182
Weakly Supervised Region Proposal Network and Object Detection
Tang, P., Wang, X., Wang, A., Yan, Y., Liu, W., Huang, J., & Yuille, A. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01252-6_22 · Cited by 202
“A previous work [17] has shown that the quality of the proposals has great influence on the performance of fully supervised object detection (i.”
BackgroundBenchmarking and Error Diagnosis in Multi-instance Pose Estimation
Ronchi, M. R., & Perona, P. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.48 · Cited by 182
“AR is defined as the maximum recall given a fixed number of detections per image [20].”
BackgroundGuirado, E., Tabik, S., Alcaraz-Segura, D., Cabello, J., & Herrera, F. (2017) · Remote Sensing · doi:10.3390/rs9121220 · Cited by 176
“• Detection proposals are techniques that employ different selection criteria to reduce the number of candidate windows, thereby avoiding the exhaustive sliding window search ([48]).”
MethodDetection of stored-grain insects using deep learning
Shen, Y., Zhou, H., Li, J., Jian, F., & Jayas, D. S. (2018) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2017.11.039 · Cited by 186
Visual Sentiment Prediction Based on Automatic Discovery of Affective Regions
Yang, J., She, D., Sun, M., Cheng, M.-M., Rosin, P. L., & Wang, L. (2018) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2018.2803520 · Cited by 190
“According to [59], [60], EdgeBoxes [57] and BING [61] are faster than methods such as Selective Search [62] and Objectness [63], while EdgeBoxes achieves better quality”
MethodEnd-to-End Airplane Detection Using Transfer Learning in Remote Sensing Images
Chen, Z., Zhang, T., & Ouyang, C. (2018) · Remote Sensing · doi:10.3390/rs10010139 · Cited by 190
“Hosang et al. [18] analyzed the performance of these region proposal algorithms in detail and found that these algorithms have low repeatability; they are not robust to noise and disturbance.”
BackgroundComposited FishNet: Fish Detection and Species Recognition From Low-Quality Underwater Videos
Zhao, Z., Liu, Y., Sun, X., Liu, J., Yang, X., & Zhou, C. (2021) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3074738 · Cited by 180
“The AR is related to the metric of the same name used in proposal evaluation [44] but calculated by category.”
MethodHu, J.-W., Zheng, B.-Y., Wang, C., Zhao, C.-H., Hou, X.-L., Pan, Q., & Xu, Z. (2020) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.1900518 · Cited by 175
RT3D: Real-Time 3-D Vehicle Detection in LiDAR Point Cloud for Autonomous Driving
Zeng, Y., Hu, Y., Liu, S., Ye, J., Han, Y., Li, X., & Sun, N. (2018) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2018.2852843 · Cited by 183
“We select 25% of the ROIs as positive to balance the distribution of examples, which helps to train a more accurate classifier [22].”
MethodReal-Time, Cloud-Based Object Detection for Unmanned Aerial Vehicles
Lee, J., Wang, J., Crandall, D., Sabanovic, S., & Fox, G. (2017) · 2017 First IEEE International Conference on Robotic Computing (IRC) · doi:10.1109/irc.2017.77 · Cited by 165
“Several object proposal methods have been recently proposed, each with strengths and weaknesses [27].”
BackgroundMethodMultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving
Teichmann, M., Weber, M., Zöllner, M., Cipolla, R., & Urtasun, R. (2016) · arXiv (Cornell University) · doi:10.17863/cam.26778 · Cited by 65
DeepBox: Learning Objectness with Convolutional Networks
Kuo, W., Hariharan, B., & Malik, J. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.285 · Cited by 176
“Bottom-up proposals [1] have two major advantages: one, by reducing the search space, they allow the usage of more sophisticated recognition machinery, and two, they localize the object well, taking this burden off the classifier.”
BackgroundMethodUrbonas, A., Raudonis, V., Maskeliūnas, R., & Damaševičius, R. (2019) · Applied Sciences · doi:10.3390/app9224898 · Cited by 158
“Using this method, the number of search regions is reduced considerably [22].”
MethodFew-Example Object Detection with Model Communication
Dong, X., Zheng, L., Ma, F., Yang, Y., & Meng, D. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2844853 · Cited by 167
“Some other algorithms [67], [68] can generate more robust and high-quality detection proposals.”
BackgroundRecurrent Pixel Embedding for Instance Grouping
Kong, S., & Fowlkes, C. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00940 · Cited by 156
From Handcrafted to Deep Features for Pedestrian Detection: A Survey
Cao, J., Pang, Y., Xie, J., Khan, F. S., & Shao, L. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3076733 · Cited by 160
“Common strategies include sliding-window methods [38], [42], [186], particle-window methods [60], [147], objectness methods [28], [69], [184], [266], and region proposal networks [11], [164], [190].”
MethodCascade RPN: Delving into High-Quality Region Proposal Network with Adaptive ConvolutionPreprint
Vu, T., Jang, H., Pham, T. X., & Yoo, C. D. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.06720 · Cited by 216
LocNet: Improving Localization Accuracy for Object Detection
Gidaris, S., & Komodakis, N. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.92 · Cited by 154
“[15] that includes a sliding window implementation inspired by BING [4, 38]). model ( Bbox reg.”
BackgroundMethodResultInfluentialSCNet: Learning Semantic Correspondence
Han, K., de Rezende, R. S., Ham, B., Wong, K.-Y. K., Cho, M., Schmid, C., & Ponce, J. (2017) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.203 · Cited by 168
“US and SW are extracted using the work of [16], and SW is similar to regular grid sampling used in other popular methods [20, 26, 33].”
BackgroundMethodPixel-Level Encoding and Depth Layering for Instance-Level Semantic Labeling
Uhrig, J., Cordts, M., Franke, U., & Brox, T. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-45886-1_2 · Cited by 147
DETReg: Unsupervised Pretraining with Region Priors for Object Detection
Bar, A., Wang, X., Kantorov, V., Reed, C. J., Herzig, R., Chechik, G., Rohrbach, A., Darrell, T., & Globerson, A. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr52688.2022.01420 · Cited by 135
Crafting GBD-Net for Object Detection
Zeng, X., Ouyang, W., Yan, J., Li, H., Xiao, T., Wang, K., Liu, Y., Zhou, Y., Yang, B., Wang, Z., et al. (2018) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2745563 · Cited by 142
“Please refer more details of the proposal average recall to [48].”
BackgroundUnified Vision‐Based Methodology for Simultaneous Concrete Defect Detection and Geolocalization
Li, R., Yuan, Y., Zhang, W., & Yuan, Y. (2018) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.12351 · Cited by 141
“5 and 1, which has been shown to be highly correlated with the performance of a detector (Hosang et al., 2016).”
BackgroundMethodInfluentialConvolutional Oriented Boundaries
Maninis, K.-K., Pont-Tuset, J., Arbeláez, P., & Van Gool, L. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46448-0_35 · Cited by 136
“Figure 9 shows the average recall [42] with respect to the number of object proposals.”
BackgroundFast Deep Vehicle Detection in Aerial Images
Sommer, L. W., Schuchert, T., & Beyerer, J. (2017) · 2017 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2017.41 · Cited by 139
“In general, object proposals methods can be categorized into grouping methods, window scoring methods and deep learning methods [11].”
MethodInfluentialShip detection with deep learning: a survey
Er, M. J., Zhang, Y., Chen, J., & Gao, W. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-023-10455-x · Cited by 141
“In addition to AP, the average recall (AR) (Hosang et al. 2015) between IoU 0.5 to 1 was introduced as a novel metric, which rewards both high recall and good localization (6) thr B gt = min 0.5, wh ( w + 10 )( h + 10 ) and correlates surprisingly well with detection performance.”
BackgroundInfluentialLiu, Z., Yang, C. F., Huang, J. J., Liu, S., Zhuo, Y., & Lu, X. (2020) · Future Generation Computer Systems · doi:10.1016/j.future.2020.08.015 · Cited by 143
A Survey on Deep-Learning-Based LiDAR 3D Object Detection for Autonomous Driving
Alaba, S. Y., & Ball, J. E. (2022) · Sensors · doi:10.3390/s22249577 · Cited by 163
“The authors use an IOU [98] to match the F-PointNet [10] 3D and BEV 2D bounding box.”
MethodInfluentialDeepProposal: Hunting Objects by Cascading Deep Convolutional Layers
Ghodrati, A., Diba, A., Pedersoli, M., Tuytelaars, T., & Gool, L. V. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.296 · Cited by 116
Pan, X., & Yang, T. Y. (2020) · Computer-Aided Civil and Infrastructure Engineering · doi:10.1111/mice.12549 · Cited by 137
Huang, X., Zhang, B., Perrie, W. A., Lu, Y., & Wang, C. (2022) · Marine Pollution Bulletin · doi:10.1016/j.marpolbul.2022.113666 · Cited by 120
Proposal Flow: Semantic Correspondences from Object Proposals
Ham, B., Cho, M., Schmid, C., & Ponce, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2017.2724510 · Cited by 146
TextProposals: A text-specific selective search algorithm for word spotting in the wild
Gómez, L., & Karatzas, D. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2017.04.027 · Cited by 107
Detecting the shuttlecock for a badminton robot: A YOLO based approach
Cao, Z., Liao, T., Song, W. W., Chen, Z., & Li, C. (2020) · Expert Systems with Applications · doi:10.1016/j.eswa.2020.113833 · Cited by 125
Track and Segment: An Iterative Unsupervised Approach for Video Object Proposals
Xiao, F., & Lee, Y. J. (2016) · 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/cvpr.2016.107 · Cited by 113
“Object proposals are beneficial in two main aspects: (1) Computation: compared to sliding window detection, they greatly reduce the number of regions in an image that must be considered (from potentially millions to thousands); and (2) Recognition accuracy: they tend to reduce non-object regions that would otherwise r…”
BackgroundTree-Structured Reinforcement Learning for Sequential Object LocalizationPreprint
Jie, Z., Liang, X., Feng, J., Jin, X., Lu, W. F., & Yan, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1703.02710 · Cited by 118
Ham, B., Cho, M., Schmid, C., & Ponce, J. (2016) · 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/cvpr.2016.378 · Cited by 123
“We use the PF benchmark in this section to compare three variants of proposal flow using different matching algorithms (NAM, PHM, LOM), combined with various object proposals [1, 24, 40, 51, 58], and features [11, 30, 31].”
Influential
Metadata
- Authors
- Hosang, J., Benenson, R., Dollar, P., & Schiele, B.
- Affiliations
- Max Planck Institute for Informatics (DE)
- ORCID
- Schiele, Bernt
- Venue
- IEEE Transactions on Pattern Analysis and Machine Intelligence
- Published
- 2016-04-01
- Volume
- 38
- Issue
- 4
- Pages
- 814-830
- ISSN
- 0162-8828 (ISSN-L)
- 2160-9292
- OpenAlex
- W1958328135
- Cited by
- 727
- References
- 68
- Citations by year
- 2026: 9 · 2025: 27 · 2024: 34 · 2023: 36 · 2022: 57 · 2021: 58 · 2020: 63 · 2019: 95 · 2018: 93 · 2017: 126 · 2016: 98
- FWCI
- 100.45
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion)
- PubMed (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- MPG.PuRe (Max Planck Society) (submittedVersion, other-oa)
- MPG.PuRe (Max Planck Society) (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 › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- object detection, PASCAL dataset, ImageNet, MS COCO, DPM, R-CNN, Fast R-CNN, sliding window search
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.