|
gptkbp:instanceOf
|
gptkb:model
|
|
gptkbp:application
|
augmented reality
autonomous vehicles
computer vision
robotics
medical imaging
content creation
object detection
image editing
|
|
gptkbp:architecture
|
transformer-based
|
|
gptkbp:canSegment
|
any object in an image
|
|
gptkbp:citation
|
high (rapidly increasing since 2023)
|
|
gptkbp:demoURL
|
https://segment-anything.com/
|
|
gptkbp:developedBy
|
gptkb:Meta_AI
|
|
gptkbp:fullName
|
gptkb:Segment_Anything_Model
|
|
gptkbp:hasDemo
|
true
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
SAM model
|
|
gptkbp:influencedBy
|
large-scale pretraining
prompt engineering
transformer architectures
|
|
gptkbp:input
|
gptkb:illustrator
|
|
gptkbp:language
|
gptkb:Python
|
|
gptkbp:license
|
Apache 2.0
|
|
gptkbp:memiliki_tugas
|
image segmentation
|
|
gptkbp:notableFor
|
scalability
zero-shot segmentation
generalization to unseen objects
interactive segmentation
|
|
gptkbp:notablePublication
|
gptkb:Segment_Anything
https://arxiv.org/abs/2304.02643
|
|
gptkbp:numberOfImagesTrainedOn
|
11 million
|
|
gptkbp:numberOfMasksTrainedOn
|
1.1 billion
|
|
gptkbp:openSource
|
true
|
|
gptkbp:organization
|
gptkb:Meta_Platforms
|
|
gptkbp:output
|
segmentation mask
|
|
gptkbp:platform
|
GPU recommended
|
|
gptkbp:promptType
|
gptkb:text
boxes
points
|
|
gptkbp:relatedTo
|
gptkb:CLIP
gptkb:Mask_R-CNN
gptkb:ViT
gptkb:DINOv2
foundation models
|
|
gptkbp:releaseYear
|
2023
|
|
gptkbp:repository
|
https://github.com/facebookresearch/segment-anything
|
|
gptkbp:trainer
|
gptkb:SA-1B_dataset
|
|
gptkbp:type
|
foundation model
|
|
gptkbp:uses
|
promptable segmentation
|
|
gptkbp:bfsParent
|
gptkb:ADDIE_model
|
|
gptkbp:bfsLayer
|
6
|