The algorithm is based on the more biologically plausible alternatives of the backpropagation (BP): direct feedback alignment (DFA) and feedback alignment (FA), which use random matrices to propagate error.
We propose a modification that is 30% faster than the flip test-time augmentation and achieves the same results for CIFAR-100.
To this end, we introduce a new benchmarking dataset, consisting of 49426 images (top-view and street-view) of 9674 buildings.
State-of-the-art deep learning methods for image processing are evolving into increasingly complex meta-architectures with a growing number of modules.
We evaluate different pasting augmentation strategies, and ultimately, we achieve 9. 7\% relative improvement on the instance segmentation and 7. 1\% on the object detection of small objects, compared to the current state of the art method on
Many machine vision applications, such as semantic segmentation and depth prediction, require predictions for every pixel of the input image.
We present a neural network model - based on CNNs, RNNs and a novel attention mechanism - which achieves 84. 2% accuracy on the challenging French Street Name Signs (FSNS) dataset, significantly outperforming the previous state of the art (Smith'16), which achieved 72. 46%.
Ranked #1 on Optical Character Recognition on FSNS - Test
We propose a new method for semantic instance segmentation, by first computing how likely two pixels are to belong to the same object, and then by grouping similar pixels together.
On the opposite end in which accuracy is critical, we present a detector that achieves state-of-the-art performance measured on the COCO detection task.
Ranked #174 on Object Detection on COCO test-dev (using extra training data)
Convolutional networks are at the core of most state-of-the-art computer vision solutions for a wide variety of tasks.
Ranked #5 on Retinal OCT Disease Classification on OCT2017