Computer Vision Serengeti is currently complete! Thanks to your efforts, we hope to improve computer vision algorithms that will eventually help us process wildlife images faster. If you want to keep classifying Serengeti wildlife, please check out www.snapshotserengeti.org!

Computer Vision: Serengeti

Help us identify and position animals within hundreds of camera trap images from the Serengeti

Learn more
Get Started!

Examine Snapshot Serengeti images to mark boxes around identifiable animals and tell us about their posture and direction, as well as how "difficult" they would be to identify.

Zooniverse Talk

Chat with the research team and other volunteers!

Join in

Computer Vision: Serengeti Statistics

View more stats

Keep track of the progress you and your fellow volunteers have made on this project.

Every click counts! Join Computer Vision: Serengeti's community to complete this project and help researchers produce important results. Click "View more stats" to see even more stats.

99%
Percent complete

By the numbers

0
Volunteers
0
Classifications
0
Subjects
0
Completed subjects

Message from the researcher

Connect with the research team on Talk to learn more about this project!

About Computer Vision: Serengeti

Our friends in the MICO Project are working with the Zooniverse to automatically detect, count and identify animals in the camera trap images we collected as part of the Snapshot Serengeti project, using a combination of image analysis and machine learning. If successful, this will allow us to use volunteers' time more productively, by optimising which images you're asked to classify - and potentially classifying some images without the need for volunteer effort.

To train the systems being developed, we need more information about the the images than the original project asked you for - we need to know where in the image the animals are located, what direction they're facing, and whether they're partially obscured. The project will initially focus on just a few specific species: elephants, buffalo, warthogs, ostriches, wildebeest, gazelles, hartebeest, guinea fowl, giraffes, and zebra. These species all have distinctive features that we hope our algorithms can distinguish.

In this project we will present you with images from past seasons of Snapshot Serengeti and ask you to draw boxes around each animal in the image. Don't worry about being pixel-perfect - we have algorithms that help us combine your answers. It will also ask you to help identify which images are "difficult" to identify and which are easy.

Thank you for your help, and we hope you enjoy this chance to revisit the wildlife of the Serengeti.

Connect with Computer Vision: Serengeti