Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Sign Up
Cited by 1,935

The DOI matches its Crossref record; 24 of the 24 references it cites were found in a registry or an index.

Cascade R-CNN: High Quality Object Detection and Instance Segmentation

Open access
Cai, Zhaowei, Vasconcelos, Nuno·2021·IEEE Transactions on Pattern Analysis and Machine Intelligence
DOI 10.1109/tpami.2019.2956516
Open this paper in CoChat — free account

Abstract

In object detection, the intersection over union (IoU) threshold is frequently used to define positives/negatives. The threshold used to train a detector defines its quality. While the commonly used threshold of 0.5 leads to noisy (low-quality) detections, detection performance frequently degrades for larger thresholds. This paradox of high-quality detection has two causes: 1) overfitting, due to vanishing positive samples for large thresholds, and 2) inference-time quality mismatch between detector and test hypotheses. A multi-stage object detection architecture, the Cascade R-CNN, composed of a sequence of detectors trained with increasing IoU thresholds, is proposed to address these problems. The detectors are trained sequentially, using the output of a detector as training set for the next. This resampling progressively improves hypotheses quality, guaranteeing a positive training set of equivalent size for all detectors and minimizing overfitting. The same cascade is applied at inference, to eliminate quality mismatches between hypotheses and detectors. An implementation of the Cascade R-CNN without bells or whistles achieves state-of-the-art performance on the COCO dataset, and significantly improves high-quality detection on generic and specific object datasets, including VOC, KITTI, CityPerson, and WiderFace. Finally, the Cascade R-CNN is generalized to instance segmentation, with nontrivial improvements over the Mask R-CNN.

Overview

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

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

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

Integrity signals

Registry

Crossref · 10.1109/tpami.2019.2956516

Cites

24 of 24 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

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

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

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

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

Research works printed but not linked: 43

  1. liu, 2016, SSD: Single shot multibox detector, Proc Eur Conf Comput Vis, 21
  2. 10.1109/CVPR.2018.00913 · DOI · Paper page
  3. 10.1109/CVPR.2015.7299170 · DOI · Paper page
  4. law, 2018, Cornernet: Detecting objects as paired keypoints, Proc Eur Conf Comput Vis, 765
  5. jiang, 2018, Acquisition of localization confidence for accurate object detection, Proc Eur Conf Comput Vis, 816
  6. 10.1145/2647868.2654889 · DOI · Paper page
  7. lin, 2014, Microsoft COCO: Common objects in context, Proc Eur Conf Comput Vis, 740
  8. li, 2018, Detnet: Design backbone for object detection, Proc Eur Conf Comput Vis, 339
  9. 10.1109/CVPR.2016.596 · DOI · Paper page
  10. zagoruyko, 2016, A multipath network for object detection, Proc BMVC
  11. 10.1109/ICCV.2015.305 · DOI
  12. 10.1109/CVPR.2017.474 · DOI · Paper page
  13. 10.1109/CVPR.2018.00442 · DOI · Paper page
  14. 10.1109/CVPR.2018.00378 · DOI · Paper page
  15. 10.1109/CVPR.2014.81 · DOI · Paper page
  16. han, 2015, Learning both weights and connections for efficient neural network, Proc Adv Neural Inf Process Syst, 1135
  17. 10.1109/ICCV.2017.322 · DOI · Paper page
  18. 10.1109/ICCV.2019.00502 · DOI · Paper page
  19. he, 2014, Spatial pyramid pooling in deep convolutional networks for visual recognition, Proc Eur Conf Comput Vis, 346
  20. simonyan, 2015, Very deep convolutional networks for large-scale image recognition, Proc ICLR
  21. 10.1109/CVPR.2016.650 · DOI
  22. 10.1109/ICCVW.2013.126 · DOI
  23. 10.1109/CVPR.2013.75 · DOI · Paper page
  24. 10.1109/CVPR.2017.634 · DOI · Paper page
  25. wu, 2018, Group normalization, Proc ECCV, 3
  26. wu, 2018, Single-shot bidirectional pyramid networks for high-quality object detection, CoRR
  27. 10.1023/B:VISI.0000013087.49260.fb · DOI · Paper page
  28. 10.1007/s11263-013-0620-5 · DOI · Paper page
  29. 10.1109/TPAMI.2011.155 · DOI · Paper page
  30. elkan, 2001, The foundations of cost-sensitive learning, Proc 17th Int Joint Conf Artif Intell, 973
  31. liu, 2018, Learning efficient single-stage pedestrian detectors by asymptotic localization fitting, Proc Eur Conf Comput Vis, 643
  32. 10.1007/3-540-59119-2_166 · DOI · Paper page
  33. gidaris, 2016, Attend refine repeat: Active box proposal generation via in-out localization, Proc BMVC
  34. 10.1109/CVPR.2016.92 · DOI · Paper page
  35. cao, 2012, Face alignment by explicit shape regression, Proc IEEE Conf Comput Vis Pattern Recognit, 2887
  36. 10.1109/CVPR.2016.343 · DOI · Paper page
  37. dai, 2016, R-FCN: Object detection via region-based fully convolutional networks, Proc Adv Neural Inf Process Syst, 379
  38. 10.1109/CVPR.2016.89 · DOI · Paper page
  39. 10.1109/CVPR.2010.5540094 · DOI · Paper page
  40. 10.1109/TPAMI.2011.281 · DOI
  41. 10.1109/CVPR.2016.260 · DOI
  42. 10.1109/TPAMI.2010.71 · DOI
  43. 10.1109/ICCV.2017.214 · DOI

Cited by

See all 1,935

Metadata

Authors
Cai, Z., & Vasconcelos, N.
Venue
IEEE Transactions on Pattern Analysis and Machine Intelligence
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
2021-05-01
Volume
43
Issue
5
Pages
1483-1498
ISSN
  • 0162-8828 (ISSN-L)
  • 2160-9292
  • 1939-3539
DOI
10.1109/tpami.2019.2956516
OpenAlex
W2989676862
Cited by
1,935
References
24
FWCI
53.63
Type
journal article
Language
en
Open access
green
Licence
Funders
NSF (IIS-1546305, IIS-1637941)

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.