Sequence related to image processing
Vision

Vision Models Get Some Attention

Self-attention is a key element in state-of-the-art language models, but it struggles to process images because its memory requirement rises rapidly with the size of the input. New research addresses the issue with a simple twist on a convolutional neural network.
2 min read
Sequence showing a training step that uses different perspectives of the same patient to enhance unsupervised pretraining
Vision

Same Patient, Different Views

When you lack labeled training data, pretraining a model on unlabeled data can compensate. New research pretrained a model three times to boost performance on a medical imaging task.
2 min read
Neighborhood being monitored by AI-powered cameras
Vision

Partners in Surveillance

Police are increasingly able to track motor vehicles throughout the U.S. using a network of AI-powered cameras — many owned by civilians. Flock, which sells automatic license plate readers is encouraging enforcers to use its network to monitor cars and trucks outside their jurisdiction.
2 min read
Graphs and data related to ReLabel, a technique that labels any random crop of any image.
Vision

Good Labels for Cropped Images

In training an image recognition model, it’s not uncommon to augment the data by cropping original images randomly. But if an image contains several objects, a cropped version may no longer match its label. Researchers developed a way to make sure random crops are labeled properly.
2 min read
Neural Body, a procedure that generates novel views of a single human character, working
Vision

Seeing People From a New Angle

The University of Hong Kong, and Cornell University to create Neural Body, a procedure that generates novel views of a single human character based on shots from only a few angles.
2 min read
Graphs and data related to ImageNet performance
Vision

ImageNet Performance: No Panacea

It’s commonly assumed that models pretrained to achieve high performance on ImageNet will perform better on other visual tasks after fine-tuning. But is it always true? A new study reached surprising conclusions.
2 min read
Graph showing system that examines X-ray images to predict which Covid-19 patients are at greatest risk of decline
Vision

Covid-19 Triage

The pandemic has pushed hospitals to their limits. A new machine learning system could help doctors make sure the most severe cases get timely, appropriate care.
2 min read
Person in wheelchair, person in side profile, person wearing a hoodie
Vision

Human Disabilities Baffle Algorithms

Facebook’s content moderation algorithms block many advertisements aimed at disabled people. The social media platform’s automated systems regularly reject ads for clothing designed for people with physical disabilities.
2 min read
System Oscar+ working
Vision

Sharper Eyes For Vision+Language

Models that interpret the interplay of words and images tend to be trained on richer bodies of text than images. Recent research worked toward giving such models a more balanced knowledge of the two domains.
2 min read
Different data related to the phenomenon called underspecification
Vision

Facing Failure to Generalize

The same models trained on the same data may show the same performance in the lab, and yet respond very differently to data they haven’t seen before. New work finds this inconsistency to be pervasive.
2 min read
Art pieces with subjective commentary regarding their emotional impact
Vision

How Art Makes AI Feel

An automated art critic spells out the emotional impact of images. Led by Panos Achlioptas, researchers at Ecole Polytechnique, King Abdullah University, and Stanford University trained a deep learning system to generate subjective interpretations of art.
2 min read
Graphs and data related to visualized tokens (or vokens)
Vision

Better Language Through Vision

For children, associating a word with a picture that illustrates it helps them learn the word’s meaning. Research aims to do something similar for machine learning models. Researchers improved a BERT model’s performance on some language tasks by training it on a large dataset of image-word pairs.
2 min read
Series of images and graphs related to cancer detection
Vision

Shortcut to Cancer Treatment

Doctors who treat breast cancer typically use a quick, inexpensive tumor-tissue stain test to diagnose the illness and a slower, more costly one to determine treatment. A new neural network could help doctors to go straight from diagnosis to treatment.
2 min read
Examples of InstaHide scrambling images
Vision

A Privacy Threat Revealed

With access to a trained model, an attacker can use a reconstruction attack to approximate its training data. A method called InstaHide recently won acclaim for promising to make such examples unrecognizable to human eyes while retaining their utility for training.
2 min read
Facebook service describing a photo on Instagram
Vision

Every Picture Tells a Story

Facebook expanded a system of vision, language, and speech models designed to open the social network to users who are visually impaired. A Facebook service that describes photos in a synthesized voice now recognizes 1,200 visual concepts — 10 times more than the previous version.
2 min read

Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox