Every species. Every site. Every day.
Every bird that enters the frame gets a name, a timestamp, and the footage attached. The feed runs every day, dawn to dusk.
Probability a species is present in any given hour.
A camera at the feeder feeds 1080p video into a dedicated edge AI device. A real-time object detector runs on every frame, followed by a fine-grained species classifier trained on North American birds.
A visit is recorded when a bird is detected for at least three consecutive frames; the visit closes after five seconds without detection. A bird that leaves the frame and reappears within that window is counted as one continuous visit, not two.
Detections that fail confidence and frame-stability checks are flagged low-quality and excluded from the public counts. Each day's full broadcast is preserved on YouTube — the source footage for every observation.
In parallel, a microphone at the feeder feeds audio into BirdNET — Cornell Lab's open-source bird sound classifier — which identifies birds by song or call. Audio identifications are displayed live on the broadcast and logged separately. They are not counted as visits, since a singing bird is not necessarily at the feeder.
This site monitors one elevated platform and tube feeder. The visit dataset is biased toward seed-eating passerines that visit feeders; raptors, ground foragers, nocturnal species, and aerial insectivores are systematically underrepresented in visits but may appear in the audio log.
Detections per hour, per species. Top 8 species shown.
BirdNET listens to everything calling around the feeder — including birds the camera never sees. The chart shows what was heard today, hour by hour.
Audio identification powered by BirdNET, Cornell Lab of Ornithology. CC BY-NC-SA 4.0.
Origin · Beyond
My wife pulls up Birds of Virginia on her phone every time something new lands at the feeder. Nothing wrong with that. It just means for the few seconds the bird is actually there, we're looking at a screen instead of the bird.
So I built a camera that does the identifying. The system around it keeps doing it after we go inside. Dawn to dusk, every day. Species, time, duration, and the footage each ID came from. All of it kept.
Yesterday at our feeder in Virginia, the camera logged 147 visits across six species. First bird of the day: a Chipping Sparrow at 6:04 AM. House Finches led in total screen time, forty-one minutes. A White-breasted Nuthatch didn't show until 11:21, then came back twenty-nine more times before sundown.
Almost everything we know about North American birds came from people watching them. Millions of people, over more than a century, feeding what they saw into the records the field now runs on. What we want to add is the thing those records can't easily hold. One place, watched without breaks, at the speed birds actually live, with footage rolling on every entry.
One yard is the start. The shape we're building toward is one of these in every state. Fifty fixed sites, same clock, same format, always on. One yard tells you what landed here. Fifty tells you where they're going.
Birds are dinosaurs. The line runs back 150 million years, and right now it's thinning. Three billion fewer birds on this continent than there were fifty years ago. If something that old is leaving, the least we can do is watch carefully while it's still here.
Every species. Every site. Every day.
Built in public
Or leave an email. We'll write when something happens.
Thanks. We'll be in touch.
Building something nearby? Want to host a node? [email protected]