Inception v1

E472887

Inception architecture version convolutional neural network architecture deep learning model architecture

Inception v1 is the original version of Google’s Inception deep convolutional neural network architecture, introduced for efficient and accurate image classification in the 2014 GoogLeNet model.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (1)

Label	Occurrences
Inception v1 canonical	1

Statements (47)

Predicate	Object
instanceOf	Inception architecture version ⓘ convolutional neural network architecture ⓘ deep learning model architecture ⓘ
achievedResultOn	ImageNet classification ⓘ
alsoKnownAs	GoogLeNet Inception architecture NERFINISHED ⓘ
basedOn	deep convolutional neural networks ⓘ
designedFor	image classification ⓘ
developedBy	Google NERFINISHED ⓘ Google Research NERFINISHED ⓘ
field	computer vision ⓘ deep learning ⓘ
hasComponent	1x1 convolution branch ⓘ 3x3 convolution branch ⓘ 5x5 convolution branch ⓘ concatenation of feature maps ⓘ pooling branch ⓘ
hasDesignGoal	computational efficiency ⓘ efficient use of parameters ⓘ high accuracy ⓘ reduced computational cost ⓘ
hasKeyIdea	balancing depth and width with computational budget ⓘ factorizing large convolutions into smaller ones via modules ⓘ
hasProperty	deep architecture ⓘ parameter efficiency ⓘ sparse connections ⓘ
hasSuccessor	Inception v2 NERFINISHED ⓘ Inception v3 NERFINISHED ⓘ
implementedIn	Caffe NERFINISHED ⓘ PyTorch NERFINISHED ⓘ TensorFlow NERFINISHED ⓘ
influenced	later Inception versions ⓘ many CNN architectures ⓘ
introducedInModel	GoogLeNet NERFINISHED ⓘ
introducedInPaper	Going Deeper with Convolutions NERFINISHED ⓘ
introducedInYear	2014 ⓘ
optimizedFor	ImageNet Large Scale Visual Recognition Challenge NERFINISHED ⓘ
partOf	GoogLeNet architecture NERFINISHED ⓘ
trainingDataset	ImageNet NERFINISHED ⓘ
usedFor	feature extraction ⓘ image recognition benchmarks ⓘ transfer learning ⓘ
usesConcept	1x1 convolutions ⓘ Inception module NERFINISHED ⓘ dimension reduction ⓘ multi-scale feature extraction ⓘ network-in-network ⓘ parallel convolutional paths ⓘ

Referenced by (1)

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

Inception architecture → hasVariant → Inception v1 ⓘ