The Wild Southwest

Motion triggered cameras, a.k.a. "camera traps," are a powerful tool for monitoring populations of many different wildlife species in a safe and non-invasive way. However, camera traps generate huge numbers of images that can be time-consuming to sort through and identify. Many researchers (including several other projects here on Zooniverse!) have been working to automate this process.

Our team is working to move one step further by also creating a process to automatically estimate the distance of an animal from the camera while also identifying it. This helps wildlife managers understand population density using statistics. Many other wildlife detection algorithms use the whole image; for our distance detection to work, we need to be able to identify the animal itself. We need your help to label images of wildlife with bounding boxes to help train the algorithm and give it examples of what to look for!