Rethinking the Inception Architecture for Computer Vision

Convolutional networks are at the core of most state of-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) s. 2818 - 2826
Hlavní autoři: Szegedy, Christian, Vanhoucke, Vincent, Ioffe, Sergey, Shlens, Jon, Wojna, Zbigniew
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 09.12.2016
Témata:
Benchmark testing
Computational efficiency
Computational modeling
Computer architecture
Computer vision
Convolution
Training
ISSN:1063-6919
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Convolutional networks are at the core of most state of-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend to translate to immediate quality gains for most tasks (as long as enough labeled data is provided for training), computational efficiency and low parameter count are still enabling factors for various use cases such as mobile vision and big-data scenarios. Here we are exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible by suitably factorized convolutions and aggressive regularization. We benchmark our methods on the ILSVRC 2012 classification challenge validation set demonstrate substantial gains over the state of the art: 21:2% top-1 and 5:6% top-5 error for single frame evaluation using a network with a computational cost of 5 billion multiply-adds per inference and with using less than 25 million parameters. With an ensemble of 4 models and multi-crop evaluation, we report 3:5% top-5 error and 17:3% top-1 error on the validation set and 3:6% top-5 error on the official test set.
ISSN:1063-6919
DOI:10.1109/CVPR.2016.308