DeepMind Unlocks Web-Scale Training for Open-World Detection

In a new paper Scaling Open-Vocabulary Object Detection, a DeepMind research team introduces OWLv2 model, an optimized architecture with improved training efficiency and applies and OWL-ST self-training recipe to the proposed OWLv2 to substantially improves detection performance, achieving state-of-the-art result on open-vocabulary detection task.

ai architecture art artificial intelligence computer vision & graphics deepmind deepmind research deep-neural-networks detection efficiency machine learning machine learning & data science ml object-detection paper performance recipe research research team scale scaling self-training state team technology training web world