ImageNet CNN

E899061

convolutional neural network model deep learning model

ImageNet CNN is a convolutional neural network model trained on the large-scale ImageNet dataset, commonly used as a powerful pretrained feature extractor for various computer vision tasks.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (1)

Label	Occurrences
ImageNet CNN canonical	1

Statements (48)

Predicate	Object
instanceOf	convolutional neural network model ⓘ deep learning model ⓘ
appliedTo	fine-grained classification ⓘ image retrieval ⓘ object detection ⓘ semantic segmentation ⓘ visual recognition ⓘ
associatedWith	ImageNet Large Scale Visual Recognition Challenge NERFINISHED ⓘ
basedOn	convolutional neural networks ⓘ
commonlyUsedAs	feature extractor ⓘ initialization for downstream models ⓘ pretrained backbone ⓘ
enables	faster convergence in downstream tasks ⓘ improved accuracy on small datasets ⓘ transfer of visual representations ⓘ
evaluationMetric	top-1 accuracy ⓘ top-5 accuracy ⓘ
hasAdvantage	captures generic low-level and mid-level visual patterns ⓘ reduces need for large labeled datasets in downstream tasks ⓘ
hasComponent	convolutional layers ⓘ fully connected layers ⓘ nonlinear activation functions ⓘ pooling layers ⓘ
hasDomain	artificial intelligence ⓘ computer vision ⓘ machine learning ⓘ
hasProperty	general-purpose visual features ⓘ hierarchical feature representations ⓘ high-dimensional feature embeddings ⓘ large-scale pretraining ⓘ learned visual features ⓘ supervised learning ⓘ
hasTrainingObjective	image classification on ImageNet ⓘ
inputType	RGB images ⓘ natural images ⓘ
outputType	class probabilities ⓘ feature vectors ⓘ
representationType	deep visual features ⓘ
trainedOn	ImageNet dataset NERFINISHED ⓘ
trainedWith	backpropagation ⓘ data augmentation ⓘ stochastic gradient descent ⓘ
usedFor	computer vision tasks ⓘ feature extraction ⓘ image classification ⓘ transfer learning ⓘ
usedIn	academic research ⓘ industrial computer vision applications ⓘ

Referenced by (1)

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

Show and Tell: A Neural Image Caption Generator → usesPretrainedModel → ImageNet CNN ⓘ