Photobooth mode also detects when people taking the selfie have good expressions with their eyes open to automatically click the photo
Google Pixel 3 is receiving a new update to its astounding cameras that will click photos automatically when the subject is kissing. The new update is being rolled out to the Photobooth mode available in Pixel 3’s camera app. With the use of AI, Pixel 3 cameras will detect movement of the subject’s lips and the area around it and automatically shoot a photograph.
In a blog post, Google has noted that Pixel 3 cameras use AI to recognise five facial expressions – smiles, tongue-out, kissy/duck face, puffy-cheeks, and surprise. When the subject is delivering an expression from the suggested list, Pixel 3 triggers the wide-angle selfie camera to take a shutter-free photograph.
Google says it has leveraged the technology used for Google Clips add-on camera clamps to develop the automatic picture capturing for the Pixel 3. The kiss-detection technology used by Photobooth is like the Image Content Model that came with Google Clips but with enhanced fine-tuning. Google notes that it has “trained a neural network” to distinguish between the facial expressions.
The kiss detection works when two subjects are kissing each other, but many users have reported the Photobooth mode triggers the camera even when they kissed the mirror. “The weighting is important, for example, to emphasize the expressions in the foreground, rather than the background,” notes Google.
The Photobooth mode also detects when people taking the selfie have good expressions with their eyes open to automatically click the photo. There is a new bar at the edge called ‘Intelligence Indicator’ that increases in length as the quality of the shot increases. This is very similar to the Best Shot feature available on Galaxy S10 cameras. The four “distinct ranges” of the bar length, as noted in the post, are – “(1) no faces clearly seen, (2) faces seen but not paying attention to the camera, (3) faces paying attention but not making key expressions, and (4) faces paying attention with key expressions.”