This facial recognition app tells one panda from another

Xinhua
Updated: May 18, 2019
A giant panda eats bamboo at the Beijing Zoo in Beijing, China, May 6, 2019. [Photo/IC]

CHENGDU -- Having trouble discerning one furry panda from another? A facial recognition app will make it easy for you.

The app is developed by the China Conservation and Research Center for Giant Pandas along with researchers in Singapore Nanyang Technological university and Sichuan Normal University.

The image analysis research kicked off in 2017. A database now contains about 120,000 images and 10,000 video clips of giant pandas.

Close to 10,000 panda pictures have been analyzed, marked and annotated.

Using the database, researchers are able to carry out automatic facial recognition on panda faces to tell one animal from another, the center said.

Panda researchers hope algorithms and AI technology will help them analyze data on pandas both in captivity and the wild.

Two pandas are shown at a zoo in Guangzhou, South China's Guangdong province, May 17, 2018. [Photo/IC]

"The app and database will help us gather more precise and well-rounded data on the population, distribution, ages, gender ratio, birth and deaths of wild pandas, who live in deep mountains and are hard to track," said Chen Peng, a researcher with the base who co-authored a paper on "Giant Panda Face Recognition Using Small Database."

"It will definitely help us improve efficiency and effectiveness in conservation and management of the animals," Chen said.

China has carried out four scientific field research project of giant pandas in the wild.

The giant panda was scientifically discovered 150 years ago and named in the city of Ya'an, Sichuan. It remains one of the world's most endangered species.

The number of captive pandas was 548 globally as of November last year. Fewer than 2,000 pandas live in the wild, mostly in the provinces of Sichuan and Shaanxi.

If you have any problems with this article, please contact us at app@chinadaily.com.cn and we'll immediately get back to you.