Computer vision
Teaching machines to interpret and reason about the visual world, from object recognition to medical imaging.
Project
Eye Control
By taking a broad view of the component parts of eye tracking – hardware, driver interface, API surface, AAC and computer control – we can find new ways to empower people who need gaze-tracking technology…
Project
Project DeepEyes
Project DeepEyes aims to develop no cost and high-quality eye tracking for every person with motor-neuron disabilities.
Video
Microsoft Vision Model ResNet-50: Pretrained vision model built with web-scale data
Pretrained computer vision models, combined with transfer learning, can dramatically bring down the cost and time it takes to build a model that performs a vision task, such as image classification, object detection, and image…
Microsoft Research Blog
HEXA: Self-supervised pretraining with hard examples improves visual representations
Humans perceive the world through observing a large number of visual scenes around us and then effectively generalizing—in other words, interpreting and identifying scenes they haven’t encountered before—without heavily relying on labeled annotations for every…
