AirPods camera code reveals image capture resolution, status light, person detection, and more

Apple is working on camera-equipped AirPods, as indicated by various reports and a demo video of them in action discovered earlier this week. The video was found by Woozad in the macOS Tahoe 26.7 release candidate, but it turns out there are even more details in the code.

Woozad forum member Mactracker has discovered a hidden macOS framework called “AccessorySensorManager” that manages sensor data from AirPods when connected to a Mac. This revealed the following information:

• AirPods designed to capture matched color images: The code indicates that the left and right earbuds have individual camera sensors that return synchronized RGB still images up to 1 megapixel with corresponding image IDs, giving the visual intelligence the same scene from two angles. This does not exclude other sensors like infrared cameras, but they are not mentioned in the code.
• AirPods support periodic image capture, but not video: Cameras take internal still images for visual intelligence to process, rather than “taking photos” or recording video. In “active” mode, the framework can request a synchronized frame pair or receive a capture rate, where still frame pairs are continuously sent at the specified rate.
• Active and passive capture modes use different resolutions: In “active” mode, presumably triggered by the user via Siri, AirPods capture a 640 x 640 image and return a processed 1024 x 1024 image. “That’s about 0.4 megapixels of sampled image data and just over 1 megapixel for the output image,” Mactracker explains. In “passive” mode, they capture a 320×320 image that can return a 320×320 or 512×512 output. “These are capture-specific resolutions, and the native resolution of the physical sensor may be larger,” adds mactracker.
• The passive mode appears intended for basic environmental awareness: The framework mentions contextual conditions such as nearby speech, changes in surrounding audio, posture changes, head rotation, and movements outside of a defined area. The code does not specify whether each condition triggers a camera capture.
• AirPods take into account continuous head movements: The framework includes lens distortion correction, dual camera calibration, and motion sensor camera calibration. It can also reject images affected by excessive movement or an obstructed view.
• Some treatments are carried out directly on the AirPods: The code mentions “peripheral inference,” or on-device detection of a person’s visibility.
• Every AirPod probably has a capture light: The framework contains code that allows it to remotely control a hardware indicator and its brightness, suggesting that AirPods have a light to notify other people when they capture still images.

It’s worth noting that the AirPods shown in the demo video were a version of the AirPods Pro 3 with cameras, which are codenamed B790. Bloomberg‘s Mark Gurman said that Apple no longer plans to release this model and is instead working on a new generation of AirPods with cameras, codenamed B798. It is therefore likely that the discovered framework is linked to the canceled model. He also points out that the results show how Apple’s camera-equipped AirPods could work, rather than a guarantee of the characteristics of the final product.

Gurman says AirPods are planned for 2027 and could potentially launch alongside the 20th anniversary iPhone.