I know I'm late, but I'm working on a contextual camera

You can't really tell what's happening but it's a 3 step process

1. The app monitors sensor inputs from your phone in realtime: (Activity, Altitude, Audio, Faces, Flash, Focus POI, Heading, Histogram, Low Light Boost, National Celebration, Network, Pitch, Property, Solar Elevation, Speed, Weather).

2. It programmatically maps sensor inputs to scenes: (Airplane, Automatic ,Automobile, Bokeh, City, Fireworks, Infant, Night Club, Selfie, Sunrise / Sunset, Foodie)

3. Then uses that scene to calibrate your phones manually camera settings (or goes to auto if it can't figure out a scene).

All of this happens in real time so all you have to do is press the shutter button and it should theoretically take the best picture possible.