In-depth look at Google+ photo update with the team that designed it
Melissa J Perenson | Published: May 17, 2013 at 21:19:34 UTC35
At its I/O developers conference a couple of days ago Google introduced various updates to its Google+ social networking platform. The main layout has changed and in the photo section "highlight photos" are selected and presented automatically, images are being "auto-enhanced" and when you upload a series of shots Google+ can automatically stitch panoramas, create an animation find the best facial expressions in group shots, similar to the features we have seen on some recent high-end smartphones.
Google+ has a whole new design, one that will change both how you can interact with the 190+ million active users of Google's social service, and how you can share your photos with them. This is no small update: Google lists 41 changes in all, starting with the fundamental design - which now appears more like the card-style layout of Pinterest, with a mutli-column Stream design and support for high-resolution images.
The emphasis on reimagining its software is consistent with this year's Google I/O theme; Google's annual developer conference focused on software and services, not hardware. Still, it's easy to see how the new design dovetails with Google's overall strategy: the new Stream layout is friendly for tablets, and the high-resolution images track with the company's Chromebook Pixel, a laptop running Google's Web-centric Chrome OS that was released earlier this year for $1,300.
In fact, those of us who view Google's motives through the rose-colored lens of a camera might go so far as to say that this move merely underscores the company's active efforts to grow the role of photos in its services. In so doing, Google sets its sights squarely on the domain of Facebook, the reigning champion of social image sharing. Facebook also took steps to shift the role of photos, as seen in last month's release of its Facebook Home app; the app is designed around images, and transforms how you view your Facebook updates and images on Android phones.
Rather than take over your home screen, Google instead provides a clean design that can be flexible in its emphasis on either text or images. Images can now run two columns, though, giving them added prominence - a feature that photographers using the service are sure to appreciate. Furthermore, Google has added a slew of editing features that go lightyears beyond what it had already integrated based on its Picnik acquisition.
Editing so often marks the difference between an image you've “made” and an image you've “taken.” The updates in Google+ reflect the company's acquisition of Nik Software last year; the new auto enhancement and auto awesome features comes from Nik's intelligence coupled with the power of Google's back-end servers and its Knowledge Graph, its previously text-centric database of 570 million people, places, and things, and 18 billion facts.
Why all the emphasis on pretty photos? In our increasingly visual world, the photo, not the text, tells the story. What better way to do so than via social media, which gains extra prominence via the proliferaiton of high-resolution smartphones and tools like Google's own Chromebook Pixel, which packs 2,560 x 1,700 pixel, 239 pixels per inch resolution (that tops Apple's so-called Retina Display on the MacBook Pro). Given the hardware support, it's only natural that Google would try to push the imaging along.
“This is just the beginning. I hope to see a 10x improvement in the year,” says Matt Steiner, one of the product managers behind the Google+ photo features. “We built it for everyone, everyday consumers who take pictures all the way to photo enthusiasts.”
Inside the new features
The main navigation is now up top, instead of a bar flush-left. Each post appears more like a single entry of info, and you can flip them around for more related info, or make them bigger as needed. You can opt for one, two, or three columns (though on my own Google+ page, I only saw layouts for one and two columns at the moment). Images are given more prominence, but it also means that if you don't have an image in your post, it looks out-of-place, too. Just another way that Google continues to emphasize the role of images.
Hashtags are more automated now, too. They appear at the upper right, and Google will add up to three hashtags based on what it understands of the context of the post.
To upload new images, you can either do so as if you're starting a new post to share, or you can select photos from the left navigation panel. Once uploaded, your images are organized by the date they were captured, or by album name.
The main photos page still lets you drill down to see what's newly-uploaded, but now has a highlights tab that shows Google's picks of the best images from a large batch you've uploaded. Google does a two-pass look at your images, pulling those that are either blurry or over- or under-exposed. “Once eliminated, we try to pick a representative set of photos,” says Steiner.
Drill down deeper to view an image in detail, and you'll see the image desktop lightbox has been completely redesigned. You can see EXIF info to the right, in the main view, and navigate around your option by zooming in, viewing, sharing, editing, or tagging people. A slideshow mode kicks images into full-screen. Buried under the “More” button is an add-to-album option, and download option. The editor in this view remains unchanged; it's the existing editor from the Picnik acquisition.
Upon uploading, each image gets altered and changed based on the context of the photo. Google says it tries to be “fairly subtle” in the changes, which impact global contrast, local tonal distribution, structure and detail, focus (via vignette adjustments), noise, skin softening, and more. The fixes examine the photo for context, but also rely on the EXIF data to find cues for adjustments. Much of Google's automation is based on sensing the context of the image. It can understand a concert based on the environment of the photo, even if it can't identify who is in the photo.
“We do a context-specific edit,” explains Steiner. “We figure out what's in the photo - this is the sky, the tree, these are people in front - and then we edit those in a context sensitive way. We selectively apply the right edits. We do different things for each photo.” Google understands, from its extensive “computer vision systems” backend (also known as a machine learning systems that have the horsepower and intelligence to identify what's in an image, what a building looks like versus a face versus a landscape.
In uploading to the Google cloud, the company can edit hundreds of photos in real-time, at the same time. The edits are lossless and non-destructive, and 100 percent reversible. And 100 percent automatic - you don't have to do anything. You can later download an individual image, or export an entire auto-fixed album, and then continue using that image on your desktop.
The new auto features aren't limited to what you upload, either. You'll be able to see photos that other people take that you've been tagged in, or see photos uploaded to a Google+ event that you attended. Those images can get mixed in with your albums, and auto-enhanced accordingly.
To review Google's automatic enhancements, just open a newly-uploaded photo in the desktop lightbox. Mouse over an image, and if it's been enhanced, a sparkly icon will pop up noting that the image has been enhanced; tap that same icon to switch between your original or the enhanced version. Don't want the auto enhance on at all? Go to More>Auto Enhance to turn the feature on and off (or do so in google.com/settings/plus).
If your photo has been Auto Awesome'd, you'll see one of five Awesome icons on the top of the photo: Mix, HDR, Pano, Smile, and Motion. This, too, can be disabled in the settings menu, if you so choose. But, there's no way to force apply any of these settings - everything is done automatically, and you have no choice or control over what settings changes Google applies. As a photographer, though, you still have the option to not allow downloads of your images - a boon that distinguishes Google+ from other services.
If you upload a series of images the Smile feature cobbles together the best facial expressions from multiple shots into a single image. Google will show you the images it drew from to create its single image, just in case you want to see the source shots. Panorama stitches together photos, while Motion turns images into an animation and HDR creates a High Dyanmic Range image from a series of bracketed shots.
Google says that over the coming months, it will begin applying its Auto Enhance and Auto Awesome filters to photographs already uploaded to its service. In the meantime, if you can't wait to try it out, you can apply the auto enhancements manually to an image, or re-upload it.
With all of these high-res images bouncing around, it's no wonder that Google upped it's free storage for all data (across Google+ and Drive) to 15GB. But for high-volume photographers 15GB will go in no time. Google does at least allow unlimited free storage for photos up to 2048 pixels. Tablet watchers will recognize that number: That image size matches that of the “Retina Display” resolution of the 9.7-inch iPad, but ironically is not enough to match the 2560 pixel resolution of Google's own Nexus 10 tablet.
Sadly, Google doesn't yet offer auto-image rescaling to match that 2048 pixel resolution. That's an unfortunate omission, and is one that will likely still steer photographers back to a desktop-based workflow that involves some level of image edit and management before uploading images to Google+. And that defeats part of the goal of the auto processing this whole update offers.