Our new Sub-Saharan Object Detection Model being used to detect animals from live video streams in Africam Safari The model can now detect 13 different animals to automatically monitor and document wildlife behaviour in remote locations. This is one of the world’s first real-time remote animal detection systems that is currently being Read more…
Cumulative species counting using object detection from the new Aeyed video analytics platform has been embedded in ConservationAI. The video shows how conservationists can monitor and track biodiversity by counting distinct species using camera traps and other devices. Here Zebras are automatically identified and calculated using the platforms counting service. Read more…
Here is an example of the Sub-Saharan model detecting a Giraffe and a group of Zebras at a drinking hole.
The video shows how the Sub-Saharan African model performs on live footage. Look out for the Luxton African detection and the detection of the car in the distance. This model is running on a consumer laptop (approx. £2000) with a GTX 2070 GPU – pretty impressive 😊. ConservationAI is a Read more…
Anyone collecting images from camera traps, drones and other media can upload them to the Conservation AI site for automatic classification. Both the original and classified image are stored online and can be reviewed anywhere.