Papers › CPPE-5: Medical Personal Protective Equipment Dataset

CPPE-5: Medical Personal Protective Equipment Dataset

15 Dec 2021arXiv:2112.09569archive 2025-07-28

Rishit Dagli, Ali Mustufa Shaikh

We present a new challenging dataset, CPPE - 5 (Medical Personal Protective Equipment), with the goal to allow the study of subordinate categorization of medical personal protective equipments, which is not possible with other popular data sets that focus on broad-level categories (such as PASCAL VOC, ImageNet, Microsoft COCO, OpenImages, etc). To make it easy for models trained on this dataset to be used in practical scenarios in complex scenes, our dataset mainly contains images that show complex scenes with several objects in each scene in their natural context. The image collection for this dataset focuses on: obtaining as many non-iconic images as possible and making sure all the images are real-life images, unlike other existing datasets in this area. Our dataset includes 5 object categories (coveralls, face shields, gloves, masks, and goggles), and each image is annotated with a set of bounding boxes and positive labels. We present a detailed analysis of the dataset in comparison to other popular broad category datasets as well as datasets focusing on personal protective equipments, we also find that at present there exist no such publicly available datasets. Finally, we also analyze performance and compare model complexities on baseline and state-of-the-art models for bounding box results. Our code, data, and trained models are available at https://git.io/cppe5-dataset.

PaperPDFCode

Code

Rishit-dagli/CPPE-Dataset officialmentioned on GitHubtf 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

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Object Detection

Datasets

Introduced by this paper, per the archive.

CPPE-5

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Object Detection CPPE-5 TridentNet AP50 85.1 #1 of 16 Archive leaderboard report
Object Detection CPPE-5 TridentNet AP75 58.3 #1 of 16 Archive leaderboard report
Object Detection CPPE-5 TridentNet APL 62.6 #1 of 16 Archive leaderboard report
Object Detection CPPE-5 TridentNet APM 41.3 #1 of 16 Archive leaderboard report
Object Detection CPPE-5 TridentNet APS 42.6 #1 of 16 Archive leaderboard report
Object Detection CPPE-5 TridentNet box AP 52.9 #1 of 16 Archive leaderboard report
Object Detection CPPE-5 Empirical Attention AP50 86.5 #2 of 16 Archive leaderboard report
Object Detection CPPE-5 Empirical Attention AP75 54.1 #2 of 16 Archive leaderboard report
Object Detection CPPE-5 Empirical Attention APL 61.0 #2 of 16 Archive leaderboard report
Object Detection CPPE-5 Empirical Attention APM 43.4 #2 of 16 Archive leaderboard report
Object Detection CPPE-5 Empirical Attention APS 38.7 #2 of 16 Archive leaderboard report
Object Detection CPPE-5 Empirical Attention box AP 52.5 #2 of 16 Archive leaderboard report
Object Detection CPPE-5 Double Heads AP50 87.3 #3 of 16 Archive leaderboard report
Object Detection CPPE-5 Double Heads AP75 55.2 #3 of 16 Archive leaderboard report
Object Detection CPPE-5 Double Heads APL 60.8 #3 of 16 Archive leaderboard report
Object Detection CPPE-5 Double Heads APM 41.0 #3 of 16 Archive leaderboard report
Object Detection CPPE-5 Double Heads APS 38.6 #3 of 16 Archive leaderboard report
Object Detection CPPE-5 Double Heads box AP 52.0 #3 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable Convolutional Network AP50 87.1 #4 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable Convolutional Network AP75 55.9 #4 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable Convolutional Network APL 61.3 #4 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable Convolutional Network APM 41.4 #4 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable Convolutional Network APS 36.3 #4 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable Convolutional Network box AP 51.6 #4 of 16 Archive leaderboard report
Object Detection CPPE-5 RegNet AP50 85.3 #5 of 16 Archive leaderboard report
Object Detection CPPE-5 RegNet AP75 51.8 #5 of 16 Archive leaderboard report
Object Detection CPPE-5 RegNet APL 60.5 #5 of 16 Archive leaderboard report
Object Detection CPPE-5 RegNet APM 41.1 #5 of 16 Archive leaderboard report
Object Detection CPPE-5 RegNet APS 35.7 #5 of 16 Archive leaderboard report
Object Detection CPPE-5 RegNet box AP 51.3 #5 of 16 Archive leaderboard report
Object Detection CPPE-5 VarifocalNet AP50 82.6 #6 of 16 Archive leaderboard report
Object Detection CPPE-5 VarifocalNet AP75 56.7 #6 of 16 Archive leaderboard report
Object Detection CPPE-5 VarifocalNet APL 58.8 #6 of 16 Archive leaderboard report
Object Detection CPPE-5 VarifocalNet APM 42.1 #6 of 16 Archive leaderboard report
Object Detection CPPE-5 VarifocalNet APS 39.0 #6 of 16 Archive leaderboard report
Object Detection CPPE-5 VarifocalNet box AP 51.0 #6 of 16 Archive leaderboard report
Object Detection CPPE-5 Localization Distillation AP50 76.5 #7 of 16 Archive leaderboard report
Object Detection CPPE-5 Localization Distillation AP75 58.8 #7 of 16 Archive leaderboard report
Object Detection CPPE-5 Localization Distillation APL 59.4 #7 of 16 Archive leaderboard report
Object Detection CPPE-5 Localization Distillation APM 43.0 #7 of 16 Archive leaderboard report
Object Detection CPPE-5 Localization Distillation APS 45.8 #7 of 16 Archive leaderboard report
Object Detection CPPE-5 Localization Distillation box AP 50.9 #7 of 16 Archive leaderboard report
Object Detection CPPE-5 FSAF AP50 84.7 #8 of 16 Archive leaderboard report
Object Detection CPPE-5 FSAF AP75 48.2 #8 of 16 Archive leaderboard report
Object Detection CPPE-5 FSAF APL 56.7 #8 of 16 Archive leaderboard report
Object Detection CPPE-5 FSAF APM 39.6 #8 of 16 Archive leaderboard report
Object Detection CPPE-5 FSAF APS 45.3 #8 of 16 Archive leaderboard report
Object Detection CPPE-5 FSAF box AP 49.2 #8 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable DETR AP50 76.9 #9 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable DETR AP75 52.8 #9 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable DETR APL 53.9 #9 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable DETR APM 35.2 #9 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable DETR APS 36.4 #9 of 16 Archive leaderboard report
Object Detection CPPE-5 Deformable DETR box AP 48.0 #9 of 16 Archive leaderboard report
Object Detection CPPE-5 Grid RCNN AP50 77.9 #10 of 16 Archive leaderboard report
Object Detection CPPE-5 Grid RCNN AP75 50.6 #10 of 16 Archive leaderboard report
Object Detection CPPE-5 Grid RCNN APL 54.4 #10 of 16 Archive leaderboard report
Object Detection CPPE-5 Grid RCNN APM 37.2 #10 of 16 Archive leaderboard report
Object Detection CPPE-5 Grid RCNN APS 43.4 #10 of 16 Archive leaderboard report
Object Detection CPPE-5 Grid RCNN box AP 47.5 #10 of 16 Archive leaderboard report
Object Detection CPPE-5 FCOS AP50 79.5 #11 of 16 Archive leaderboard report
Object Detection CPPE-5 FCOS AP75 45.9 #11 of 16 Archive leaderboard report
Object Detection CPPE-5 FCOS APL 51.7 #11 of 16 Archive leaderboard report
Object Detection CPPE-5 FCOS APM 39.2 #11 of 16 Archive leaderboard report
Object Detection CPPE-5 FCOS APS 36.7 #11 of 16 Archive leaderboard report
Object Detection CPPE-5 FCOS box AP 44.4 #11 of 16 Archive leaderboard report
Object Detection CPPE-5 Faster RCNN AP50 73.8 #12 of 16 Archive leaderboard report
Object Detection CPPE-5 Faster RCNN AP75 47.8 #12 of 16 Archive leaderboard report
Object Detection CPPE-5 Faster RCNN APL 52.5 #12 of 16 Archive leaderboard report
Object Detection CPPE-5 Faster RCNN APM 34.7 #12 of 16 Archive leaderboard report
Object Detection CPPE-5 Faster RCNN APS 30.0 #12 of 16 Archive leaderboard report
Object Detection CPPE-5 Faster RCNN box AP 44.0 #12 of 16 Archive leaderboard report
Object Detection CPPE-5 Sparse RCNN AP50 69.6 #13 of 16 Archive leaderboard report
Object Detection CPPE-5 Sparse RCNN AP75 44.6 #13 of 16 Archive leaderboard report
Object Detection CPPE-5 Sparse RCNN APL 54.7 #13 of 16 Archive leaderboard report
Object Detection CPPE-5 Sparse RCNN APM 30.6 #13 of 16 Archive leaderboard report
Object Detection CPPE-5 Sparse RCNN APS 30.0 #13 of 16 Archive leaderboard report
Object Detection CPPE-5 Sparse RCNN box AP 44.0 #13 of 16 Archive leaderboard report
Object Detection CPPE-5 RepPoints AP50 75.9 #14 of 16 Archive leaderboard report
Object Detection CPPE-5 RepPoints AP75 40.1 #14 of 16 Archive leaderboard report
Object Detection CPPE-5 RepPoints APL 48.0 #14 of 16 Archive leaderboard report
Object Detection CPPE-5 RepPoints APM 36.7 #14 of 16 Archive leaderboard report
Object Detection CPPE-5 RepPoints APS 27.3 #14 of 16 Archive leaderboard report
Object Detection CPPE-5 RepPoints box AP 43.0 #14 of 16 Archive leaderboard report
Object Detection CPPE-5 YOLOv3 AP50 79.4 #15 of 16 Archive leaderboard report
Object Detection CPPE-5 YOLOv3 AP75 35.3 #15 of 16 Archive leaderboard report
Object Detection CPPE-5 YOLOv3 APL 49.0 #15 of 16 Archive leaderboard report
Object Detection CPPE-5 YOLOv3 APM 28.4 #15 of 16 Archive leaderboard report
Object Detection CPPE-5 YOLOv3 APS 23.1 #15 of 16 Archive leaderboard report
Object Detection CPPE-5 YOLOv3 box AP 38.5 #15 of 16 Archive leaderboard report
Object Detection CPPE-5 SSD AP50 57.0 #16 of 16 Archive leaderboard report
Object Detection CPPE-5 SSD AP75 24.9 #16 of 16 Archive leaderboard report
Object Detection CPPE-5 SSD APL 34.6 #16 of 16 Archive leaderboard report
Object Detection CPPE-5 SSD APM 23.1 #16 of 16 Archive leaderboard report
Object Detection CPPE-5 SSD APS 32.1 #16 of 16 Archive leaderboard report
Object Detection CPPE-5 SSD box AP 29.50 #16 of 16 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.

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