AI tech recognizes human actions from just a few example videos

Дата публикации: 26-03-2026 18:50:01

Typically, AI requires massive amounts of training data to understand complex human actions. However, in real-world scenarios, it is often difficult to secure sufficient video data for specific actions. A research team led by Jae-Pil Heo, Professor in the Department of Software at Sungkyunkwan University, has developed an AI technology that can accurately recognize new actions from only a small number of example videos. The research team focused on few-shot action recognition, which enables AI to learn and distinguish the characteristics of new actions from only a few examples.

Основное содержимое страницы с новостью.

Схожие новости

#	Наименование новости	Тональность	Информативность	Дата публикации
1	New tool identifies the sources of fake videos	0	7.43	27-07-2026
2	Mouse moves unlock realistic AI video control with no extra computing cost	7	8	22-06-2026
3	Reliably detecting and clearly explaining deepfake images	5	7	01-07-2026
4	Reconfigurable AI device preserves long-term memory while acoustic waves reset short-term signals	0	11.14	28-07-2026
5	Researchers Develop New Method for Detecting Manipulated Videos	5	8	16-07-2026
6	A hardware-software co-design can efficiently run AI on edge devices	5	7	11-04-2026
7	New AI video tool removes objects without breaking the laws of physics	0	5	07-04-2026
8	Ученые научили нейросети распознавать запрещенный к показу контент в видеопотоке	0	0	05-10-2019
9	Нейросеть научилась распознавать поведение мышей без разметки	5	7	12-07-2026
10	Ученые РФ разрабатывают алгоритмы для квантовых систем распознавания лиц на видеозаписях	0	0	13-08-2019

Классификация: Наука. Схожих патентов: 0. Схожих новостей: 10. Тональность: 0. Информативность: 7. Источник: phys.org.