Papers › Cascade R-CNN: Delving into High Quality Object Detection

Cascade R-CNN: Delving into High Quality Object Detection

3 Dec 2017CVPR 2018 6arXiv:1712.00726archive 2025-07-28

Zhaowei Cai, Nuno Vasconcelos

In object detection, an intersection over union (IoU) threshold is required to define positives and negatives. An object detector, trained with low IoU threshold, e.g. 0.5, usually produces noisy detections. However, detection performance tends to degrade with increasing the IoU thresholds. Two main factors are responsible for this: 1) overfitting during training, due to exponentially vanishing positive samples, and 2) inference-time mismatch between the IoUs for which the detector is optimal and those of the input hypotheses. A multi-stage object detection architecture, the Cascade R-CNN, is proposed to address these problems. It consists of a sequence of detectors trained with increasing IoU thresholds, to be sequentially more selective against close false positives. The detectors are trained stage by stage, leveraging the observation that the output of a detector is a good distribution for training the next higher quality detector. The resampling of progressively improved hypotheses guarantees that all detectors have a positive set of examples of equivalent size, reducing the overfitting problem. The same cascade procedure is applied at inference, enabling a closer match between the hypotheses and the detector quality of each stage. A simple implementation of the Cascade R-CNN is shown to surpass all single-model object detectors on the challenging COCO dataset. Experiments also show that the Cascade R-CNN is widely applicable across detector architectures, achieving consistent gains independently of the baseline detector strength. The code will be made available at https://github.com/zhaoweicai/cascade-rcnn.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="1712.00726")

Code

Syntology Ran 2 of 2 code samples harvested from 0 repositories linked to this paper; 0 have no recorded run. Of those that ran: 2 ran · our draft was wrong.

By repository: 2 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

zhaoweicai/cascade-rcnn officialmentioned in papertf report
makefile/frcnn mentioned on GitHub report
tensorpack/tensorpack mentioned on GitHubtf report
tsbiosky/Cascade-point-rcnn mentioned on GitHubpytorch report
open-mmlab/mmdetection pytorchApache-2.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

2 samples harvested; 2 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · our draft was wrong

Licence: 2 of the 2 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

create_logger identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · bae007593bd3dfd1 · report
create_logger identical code first harvested elsewhere ran · our draft was wrong licence of this copy not recorded · c349bc6f8ed16220 · report

Tasks

2D Object DetectionObjectObject DetectionVocal Bursts Intensity Prediction

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
2D Object Detection SARDet-100K Cascade R-CNN box mAP 51.1 #6 of 13 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) AP 13.8 #6 of 7 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) AP50 30.8 #6 of 7 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) AP75 10.5 #6 of 7 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) APm 26.6 #6 of 7 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) APs 25.5 #6 of 7 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) APt 10.6 #6 of 7 Archive leaderboard report
Object Detection AI-TOD Cascade R-CNN (ResNet-50-FPN) APvt 0.0 #6 of 7 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-101-FPN+, cascade) AP50 61.6 #150 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-101-FPN+, cascade) AP75 46.6 #150 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-101-FPN+, cascade) APL 57.4 #150 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-101-FPN+, cascade) APM 46.2 #150 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-101-FPN+, cascade) APS 23.8 #150 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-101-FPN+, cascade) box AP 42.7 #150 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-50-FPN+) AP50 59.4 #180 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-50-FPN+) AP75 43.7 #180 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-50-FPN+) APL 54.1 #180 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-50-FPN+) APM 43.7 #180 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-50-FPN+) APS 22.9 #180 of 220 Archive leaderboard report
Object Detection COCO minival Cascade R-CNN (ResNet-50-FPN+) box AP 40.3 #180 of 220 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+, cascade) AP50 62.1 #168 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+, cascade) AP75 46.3 #168 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+, cascade) APL 55.2 #168 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+, cascade) APM 45.5 #168 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+, cascade) APS 23.7 #168 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+, cascade) box mAP 42.8 #168 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) AP50 59.9 #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) AP75 44 #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) APL 52.1 #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) APM 42.7 #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) APS 22.6 #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) Hardware Burden 12G #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+, cascade) box mAP 40.6 #189 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) AP50 61.1 #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) AP75 41.9 #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) APL 49.8 #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) APM 41.8 #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) APS 21.3 #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) Hardware Burden 3G #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-101-FPN+) box mAP 38.8 #209 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) AP50 59 #222 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) AP75 39.2 #222 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) APL 46.4 #222 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) APM 38.8 #222 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) APS 20.3 #222 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) Hardware Burden 3G #222 of 225 Archive leaderboard report
Object Detection COCO test-dev Cascade R-CNN (ResNet-50-FPN+) box mAP 36.5 #222 of 225 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Introduced by this paper: Cascade Mask R-CNN, Cascade R-CNN

1x1 ConvolutionAverage PoolingBatch NormalizationBottleneck Residual BlockCascade Mask R-CNNCascade R-CNNConvolutionDense ConnectionsDropoutFPNFaster R-CNNGlobal Average PoolingGroup NormalizationGrouped ConvolutionKaiming InitializationMax PoolingPosition-Sensitive RoI PoolingR-FCNRPNReLUResNeXtResNeXt BlockResidual BlockResidual ConnectionRoIAlignRoIPoolSoftmax

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections