Today, automatic exposure control (AEC) is a standard feature on cameras. AEC automatically determines exposure settings for an image without user input. Using AEC, a camera may determine an exposure setting for the camera. However, AEC typically measures an amount of light in a field of view, with no reference to what is in the field of view. Therefore, there is a need in the art for improved AEC.
Provided are techniques for determining one or more settings (e.g., an exposure setting and/or a gain setting) for a content capture device. In some examples, to determine the one or more settings, luma values of pixels of an image from the content capture device may be identified. Objects in the image and information associated with the objects may also be identified. The information associated with the objects may be divided into categories. And, using the objects and the information, a separate weight array for each category of the information may be computed. The separate weight arrays may be combined to create a total weight array to augment the luma values. The augmented luma values may be aggregated to compute a weighted luma average for the image. Based upon a difference of the weighted luma average and a target, the one or more settings may be adjusted.
In other examples, rather than computing a separate weight array for each category of the information, the information associated with each object may be used in a separate learning based model. Combining the output of each learning based model, a total weight array may be created to augment the luma values of an image. The augmented luma values may be aggregated to compute a weighted luma average for the image. Based upon a difference of the weighted luma average and a target, the one or more settings may be adjusted.