There is publicity about the new Pixel 4’s astrophotography mode, with plenty of customers posting amazing photos of the night stars all over the social media platform. Astrophotography is an art of capturing stars & other celestial bodies, and this is not new to smartphones. The Huawei’s P30 Pro also featured this ability in March, but Google’s approach is a bit different.
Everyone has tried to photoshoot the sky in the night with a dedicated camera like a DSLR, and all are familiar with the procedure of how to shoot long exposures, where to enables the shutter of the camera to stay open for a long time than usual time to capture extra light. This works with a large sensor, but the relatively small sensor in a phone like a Pixel 4 naturally absorbs less light, so it needs a long exposure. The problematic part about too long exposing is that the Earth’s rotation starts to affect on how the sky is being captured & one will end up with star trails or some white lines in the sky as a consequence of the different positions of the stars at the start & end of the exposure. Google has compensated for this; it still allows the Pixel 4 to take insufficient light for a clear photograph. Astrophotography is made into the Night sky camera mode in the application, which is called stock camera of the Pixel & operates automatically when it detects dark conditions. A capture can take five minutes to complete, but this is not a five-minute exposure. Instead, the device receives a sequence of 16-sec exposures & stitches them together in real-time, using the device gyroscope & computational data to bring into different line exposure with the last one. Think of it the feature like an HDR+ advanced form with some of the exposures arranging highlights & others shadows.
Google has a new feature named as Semantic Segmentation that is similar to Apple’s Deep Fusion, it identifies various objects in the shot & processes them another way. This allows the PIxel 4 to sharpen the picture of the stars & also brighten the foreground, independently of each other. Besides, there are also some new improvements to the processing of noise reduction, which battle against the white pixels that appear in photographs.