All Notes

End-to-end object detection with Transformers

Traditional computer vision models typically use a complex, partly handcrafted pipeline that relies on custom layers in order to localize objects in an image and then extract features. DETR replaces this with a simpler neural network that offers a true end-to-end deep learning solution to the problem. (via @reckoningdev)

Facebook AI Blog

Data-efficient image Transformers

Image classification — the task of understanding the main content of an image — is easy for humans but hard for machines. In particular, it is challenging for convolution-free Transformers like DeiT because these systems don’t have many statistical priors about images: They typically have to “see” a lot of example images in order to learn to classify different objects. DeiT, however, can be trained effectively with 1.2 million images, rather than requiring hundreds of millions of images. (via @reckoningdev)

Facebook AI Blog

The Surprising Effectiveness of Linear Unsupervised Image-to-Image Translation

Unsupervised image-to-image translation is an inherently ill-posed problem. This paper uses linear encoder-decoder architectures for unsupervised image to image translation. (via @reckoningdev)

http://arxiv.org/abs/2007.12568v1

Robust Training with Ensemble Consensus

Since deep neural networks are over-parameterized, they can memorize noisy examples. We address such memorizing issue in the presence of annotation noise. (via @reckoningdev)

https://openreview.net/pdf?id=ryxOUTVYDH

BorderDet: Border Feature for Dense Object Detection

Dense object detectors rely on the sliding-window paradigm that predicts the object over a regular grid of image. Meanwhile, the feature maps on the point of the grid are adopted to generate… (via @reckoningdev)

https://arxiv.org/pdf/2007.11056v1.pdf

Combatting Bias in Medical AI Systems - Charles E. Kahn, Jr

Those of us who see the great potential of artificial intelligence in radiology are eager to assure that AI systems work to the benefit of all of our patients. To do so, we must be aware of possibilities for error. In quality management, a latent error is a failure that is “waiting to happen,” often due to an oversight in design or execution.

https://pubs.rsna.org/page/ai/blog/2020/7/ryai_editorsblog0715