Trending

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

Added to collectionTop 100 Deep Learning and AI Papers 2021

Computer Vision

Knowledge Discovery from a Single Image

Extrapolating from a Single Image to a Thousand Classes using Distillation

We develop a framework for training neural networks from scratch using a single image by means of knowledge distillation from a supervisedly pretrained teacher.

With this, we find top-1 accuracies of 94%/74% on cifar-10/100, 59% on imagenet and, by extending this method to audio, 84% on speechcommands.

In extensive analyses we disentangle the effect of augmentations, choice of source image and network architectures and also discover"panda neurons"in networksthat have never seen a panda.

Authors

Yuki M. Asano, Aaqib Saeed

Related Topics

Neural networks

Network architectures

Read the Paper

Read Paper

Trending

Bookmarks

◐ Recommended Follows

Julien Chaumond

More

◐ Latest Activity

More

Knowledge Discovery from a Single Image - 42Papers

Read Related Papers

A generative adversarial network for style augmentation for nuclei segmentation

Multi-Label Iterated Learning

A generative adversarial network based image transformation policy for visual self-supervision

Tagger: Deep Unsupervised Perceptual Grouping

Adversarial AutoAugment

Causality-inspired Single-source Domain Generalization for Medical Image Segmentation

A Method for Evaluating the Capacity of Generative Adversarial Networks to Reproduce High-order Spatial Context

PeCo: Perceptual Codebook for BERT Pre-training of Vision Transformers

Scaling Up Vision-Language Pre-training for Image Captioning

Effective Evaluation of Deep Active Learning on Image Classification Tasks