AI DEVELOPED TO RECOGNIZE THE FACE OF CHIMP IN THE WILD
Scientists at the University of Oxford have developed new artificial intelligence software to recognise and track the faces of individual chimpanzees in the wild.
The new software will allow researchers and wildlife conservationists to significantly cut back on time and resources spent analysing video footage, according to a research paper published in Science Advances.
“For species like chimpanzees, which have complex social lives and live for many years, getting snapshots of their behaviour from short-term field research can only tell us so much,” said Dan Schofield, researcher and DPhil student at Oxford University’s Primate Models Lab, School of Anthropology.
“By harnessing the power of machine learning to unlock large video archives, it makes it feasible to measure behaviour over the long term, for example observing how the social interactions of a group change over several generations.”
The computer model was trained using over 10 million images from Kyoto University’s Primate Research Institute (PRI) video archive of wild chimpanzees in Guinea, West Africa. The new software is the first to continuously track and recognise individuals in a wide range of poses, performing with high accuracy in difficult conditions such as low lighting, poor image quality and motion blur.
“Access to this large video archive has allowed us to use cutting-edge deep neural networks to train models at a scale that was previously not possible,” said Arsha Nagrani, co-author of the study and DPhil student at the Department of Engineering Science, University of Oxford.
“Additionally, our method differs from previous primate face-recognition software in that it can be applied to raw video footage with limited manual intervention or pre-processing, saving hours of time and resources.”
The technology has potential for many uses, including monitoring species for conservation. Although the current application focused on chimpanzees, the software provided could be applied to other species and help drive the adoption of artificial intelligence systems to solve a range of problems in the wildlife sciences.
“All our software is available open-source for the research community,” says Nagrani. “We hope that this will help researchers across other parts of the world apply the same cutting-edge techniques to their unique animal data sets. As a computer vision researcher, it is extremely satisfying to see these methods applied to solve real, challenging biodiversity problems.
“With an increasing biodiversity crisis and many of the world’s ecosystems under threat, the ability to closely monitor different species and populations using automated systems will be crucial for conservation efforts, as well as animal behaviour research,” Schofield added.
“Interdisciplinary collaborations like this have huge potential to make an impact by finding novel solutions for old problems and asking biological questions which were previously not feasible on a large scale.”
Wildlife conservation projects are increasingly using cutting-edge technology in their efforts to monitor, track and protect animals. Earlier this year, an artificial intelligence system created by Intel is being used in cameras to help detect poachers illegally entering wildlife reserves.
Intel’s software is being used in TrailGuard AI cameras that are capable remotely of object detection and image classification. The system will alert park rangers should a person or vehicle be detected, so that the rangers can tackle the poachers before they kill endangered animals.
The cameras use Intel’s neural network algorithms to help them more accurately identify poachers rather than other motion in front of the camera.
These AI cameras have been deployed in African wildlife reserves as well as throughout south-east Asia. The pencil-sized devices contain a long-life battery which can last up to 18 months without needing to be recharged.
Also, the gaps between channels in the digital TV spectrum are being used by conservationists to test technology for monitoring endangered wildlife in remote areas.
The Zoological Society of London (ZSL) and Google have tested using television white space (TVWS) to send footage from cameras in the wild, as the low frequencies in this part of the spectrum can travel long distances, making it well suited to provide low-cost connectivity to remote regions.
It is thought that these vacant frequencies could allow conservationists to accurately monitor areas such as rainforests or deserts, where it has previously been difficult to study wildlife.
As well as wildlife preservation efforts, conservationists are also investigating the possibility that scientific advances could help to save endangered species
The new software will allow researchers and wildlife conservationists to significantly cut back on time and resources spent analysing video footage, according to a research paper published in Science Advances.
“For species like chimpanzees, which have complex social lives and live for many years, getting snapshots of their behaviour from short-term field research can only tell us so much,” said Dan Schofield, researcher and DPhil student at Oxford University’s Primate Models Lab, School of Anthropology.
“By harnessing the power of machine learning to unlock large video archives, it makes it feasible to measure behaviour over the long term, for example observing how the social interactions of a group change over several generations.”
The computer model was trained using over 10 million images from Kyoto University’s Primate Research Institute (PRI) video archive of wild chimpanzees in Guinea, West Africa. The new software is the first to continuously track and recognise individuals in a wide range of poses, performing with high accuracy in difficult conditions such as low lighting, poor image quality and motion blur.
“Access to this large video archive has allowed us to use cutting-edge deep neural networks to train models at a scale that was previously not possible,” said Arsha Nagrani, co-author of the study and DPhil student at the Department of Engineering Science, University of Oxford.
“Additionally, our method differs from previous primate face-recognition software in that it can be applied to raw video footage with limited manual intervention or pre-processing, saving hours of time and resources.”
The technology has potential for many uses, including monitoring species for conservation. Although the current application focused on chimpanzees, the software provided could be applied to other species and help drive the adoption of artificial intelligence systems to solve a range of problems in the wildlife sciences.
“All our software is available open-source for the research community,” says Nagrani. “We hope that this will help researchers across other parts of the world apply the same cutting-edge techniques to their unique animal data sets. As a computer vision researcher, it is extremely satisfying to see these methods applied to solve real, challenging biodiversity problems.
“With an increasing biodiversity crisis and many of the world’s ecosystems under threat, the ability to closely monitor different species and populations using automated systems will be crucial for conservation efforts, as well as animal behaviour research,” Schofield added.
“Interdisciplinary collaborations like this have huge potential to make an impact by finding novel solutions for old problems and asking biological questions which were previously not feasible on a large scale.”
Wildlife conservation projects are increasingly using cutting-edge technology in their efforts to monitor, track and protect animals. Earlier this year, an artificial intelligence system created by Intel is being used in cameras to help detect poachers illegally entering wildlife reserves.
Intel’s software is being used in TrailGuard AI cameras that are capable remotely of object detection and image classification. The system will alert park rangers should a person or vehicle be detected, so that the rangers can tackle the poachers before they kill endangered animals.
The cameras use Intel’s neural network algorithms to help them more accurately identify poachers rather than other motion in front of the camera.
These AI cameras have been deployed in African wildlife reserves as well as throughout south-east Asia. The pencil-sized devices contain a long-life battery which can last up to 18 months without needing to be recharged.
Also, the gaps between channels in the digital TV spectrum are being used by conservationists to test technology for monitoring endangered wildlife in remote areas.
The Zoological Society of London (ZSL) and Google have tested using television white space (TVWS) to send footage from cameras in the wild, as the low frequencies in this part of the spectrum can travel long distances, making it well suited to provide low-cost connectivity to remote regions.
It is thought that these vacant frequencies could allow conservationists to accurately monitor areas such as rainforests or deserts, where it has previously been difficult to study wildlife.
As well as wildlife preservation efforts, conservationists are also investigating the possibility that scientific advances could help to save endangered species
Comments
Post a Comment
🍸🍸🍹🍹