MaskRCNN

E431009

convolutional neural network deep learning model architecture instance segmentation model object detection model

MaskRCNN is a deep learning model architecture for instance segmentation that extends Faster R-CNN by adding a branch to predict segmentation masks for individual objects in an image.

Try in SPARQL Jump to: Surface forms Disambiguation Statements Elicitation Referenced by

All labels observed (4)

Label	Occurrences
Mask R-CNN	4
Mask R-CNN (paper)	1
MaskRCNN canonical	1
RoIAlign	1

How this entity was disambiguated

This entity first appeared as the object of triple T4326004 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.

NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07

Target entity: MaskRCNN
Context triple: [torchvision, modelFamily, MaskRCNN]

A. DETR
DETR is the acronym for the former UK government Department of the Environment, Transport and the Regions, which was responsible for environmental policy, transport, and regional affairs.
B. YOLO
"YOLO" is a comedic hip-hop song and music video by The Lonely Island, featuring Adam Levine and Kendrick Lamar, that parodies the phrase "you only live once" by humorously promoting extreme caution and risk avoidance.
C. ResNet
ResNet is a deep convolutional neural network architecture known for its use of residual connections to enable very deep models and achieve state-of-the-art performance in image recognition tasks.
D. ResNeXt
ResNeXt is a deep convolutional neural network architecture that extends ResNet by using grouped convolutions and a split-transform-merge strategy to improve accuracy and efficiency in image recognition tasks.
E. CLIP
CLIP is an OpenAI model that learns joint representations of images and text, enabling tasks like zero-shot image classification and natural language-based image retrieval.
F. None of above. chosen
G. Unsure - the case is ambiguous/there is not enough information to decide.

NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07

Target entity: MaskRCNN
Target entity description: MaskRCNN is a deep learning model architecture for instance segmentation that extends Faster R-CNN by adding a branch to predict segmentation masks for individual objects in an image.

A. DETR
DETR is the acronym for the former UK government Department of the Environment, Transport and the Regions, which was responsible for environmental policy, transport, and regional affairs.
B. YOLO
"YOLO" is a comedic hip-hop song and music video by The Lonely Island, featuring Adam Levine and Kendrick Lamar, that parodies the phrase "you only live once" by humorously promoting extreme caution and risk avoidance.
C. ResNet
ResNet is a deep convolutional neural network architecture known for its use of residual connections to enable very deep models and achieve state-of-the-art performance in image recognition tasks.
D. ResNeXt
ResNeXt is a deep convolutional neural network architecture that extends ResNet by using grouped convolutions and a split-transform-merge strategy to improve accuracy and efficiency in image recognition tasks.
E. CLIP
CLIP is an OpenAI model that learns joint representations of images and text, enabling tasks like zero-shot image classification and natural language-based image retrieval.
F. None of above. chosen

Statements (50)

Predicate	Object
instanceOf	convolutional neural network ⓘ deep learning model architecture ⓘ instance segmentation model ⓘ object detection model ⓘ
application	autonomous driving ⓘ image segmentation ⓘ medical image analysis ⓘ video analysis ⓘ
basedOn	Faster R-CNN ⓘ linked to: FasterRCNN
commonlyTrainedOn	COCO dataset ⓘ linked to: COCO object detection benchmarks PASCAL VOC ⓘ
commonlyUsedBackbone	ResNeXt ⓘ ResNet-101 ⓘ linked to: ResNet ResNet-50 ⓘ linked to: ResNet
extends	Faster R-CNN ⓘ linked to: FasterRCNN
field	computer vision ⓘ deep learning ⓘ
hasComponent	RoIAlign layer ⓘ backbone network ⓘ bounding box regression head ⓘ classification head ⓘ mask prediction branch ⓘ region proposal network ⓘ
hasLossFunction	bounding box regression loss ⓘ classification loss ⓘ mask loss ⓘ
implementedIn	Detectron ⓘ Detectron2 ⓘ linked to: Detectron PyTorch ⓘ TensorFlow ⓘ
improvesUpon	Faster R-CNN ⓘ linked to: FasterRCNN
introducedBy	Georgia Gkioxari ⓘ Kaiming He ⓘ Piotr Dollár ⓘ Ross Girshick ⓘ
introducedInPaper	Mask R-CNN ⓘ linked to: MaskRCNN
outputs	per-instance segmentation ⓘ pixel-level object masks ⓘ
predicts	bounding boxes ⓘ object class labels ⓘ segmentation masks ⓘ
publishedAtConference	ICCV 2017 ⓘ linked to: IEEE International Conference on Computer Vision
solvesProblem	misalignment of RoI features ⓘ
task	instance segmentation ⓘ object detection ⓘ object localization ⓘ
uses	RoIAlign ⓘ convolutional layers ⓘ feature pyramid networks ⓘ
yearIntroduced	2017 ⓘ

How these facts were elicited

Referenced by (7)

Full triples — surface form annotated when it differs from this entity's canonical label.

torchvision → modelFamily → MaskRCNN ⓘ

subject linked to: torchvision (ecosystem)

Kaiming He → knownFor → Mask R-CNN ⓘ

linked to: MaskRCNN

Kaiming He → notableWork → Mask R-CNN (paper) ⓘ

linked to: MaskRCNN

Faster R-CNN → influenced → Mask R-CNN ⓘ

subject linked to: FasterRCNN

linked to: MaskRCNN

Mask R-CNN → introducedInPaper → Mask R-CNN ⓘ

subject linked to: MaskRCNN

linked to: MaskRCNN

KeypointRCNN → basedOn → Mask R-CNN ⓘ

linked to: MaskRCNN

KeypointRCNN → uses → RoIAlign ⓘ

linked to: MaskRCNN

All labels observed (4)

How this entity was disambiguated Show

Statements (50)

How these facts were elicited Show

Referenced by (7)

How this entity was disambiguated

How these facts were elicited