Connect with us

Camera

Cloud companies may lease your images for AI learning; what can you do about it?

Published

on

Cloud companies may lease your images for AI learning; what can you do about it?


Photo sharing site Photobucket recently revealed that it might license the images it hosts to teach AI models.

Photobucket’s recent reveal that its 13 billion images may soon be licensed to teach AI models raises questions about copyright and how to protect your images.

Cloud storage is a useful tool that allows us to back up our images and share them with others quickly. It’s not perfect, though. Constant monthly payments and the possibility of data loss or leaks turn many people away. Now, with the rise of cloud-based companies potentially using your data for AI machine learning, it begs the question: what can we do to keep our files safe and, most importantly, under our own control and private?

We learned recently that Photobucket, one of the internet’s oldest photo and video storage sites, was in talks to license user data to companies wanting to use it to train AI models. In an interview with Reuters, the company’s CEO, Ted Leonard, said, “We need to pay our bills, and this could give us the ability to continue to support free accounts.” This means granting third-party access to the 13 billion images and videos uploaded to the platform.

Changes to T&Cs

It’s possible for Photobucket to do this thanks to an update in its terms and conditions, which gives the company carte blanche to analyze and incorporate data from user images. The privacy policy states, “We, and those third parties we may contract with, may also use your images in a general and anonymous way for training artificial intelligence (“AI”) algorithms and/or machine learning models and subsequent commercial uses derived therefrom.”

Cloud companies having the ability to scan user images is nothing new. Popular services such as Dropbox, Google Drive and Adobe Cloud can all scan images, as can many others. Google and Adobe have both been accused of accessing user data to train AI, though both companies deny doing so.

“It’s possible for Photobucket to do this thanks to an update in its terms and conditions.”

Personally, I’m not comfortable with anyone other than myself and those I authorize accessing my images. That’s even more true now when there’s no clarity on how severely AI technology will impact those working in the photo industry.

Some may suggest reverting to backing up images locally to a hard drive. For many, including myself, this isn’t a realistic option. As someone who travels a lot, I find storing my files in the cloud much easier than carrying multiple hard drives whenever I’m on the move.

Finding privacy in the cloud

Thankfully, there are alternatives to the status quo and several cloud services that emphasize keeping user data private, away from the eyes of company employees and third parties. Of course, every service can update its T&Cs, but it would be bad business for those who have marketed themselves as taking user privacy seriously.

Over five years, I’ve tested and reviewed dozens of cloud storage services. The first question I ask when looking at a service is: does it offer zero-knowledge encryption? This type of encryption gives the user a unique key, which even the software developer can not access, and by extension, means they can’t grant third-party companies access to your files. This means only the user, or those they share their encryption key with, can access their data.

“The first question I ask when looking at a service is: does it offer zero-knowledge encryption?”

When considering platforms that offer zero-knowledge encryption, it can be difficult to know where’s best to back up your files. Having tried the good and the bad, there are some services I feel confident recommending to others.

The two I’m currently using are pCloud and Sync.com, as I find them the easiest to use, and the pricing, overall, is affordable. They’re both available on desktop and mobile and offer useful features for photographers and videographers, which I’ll break down in more detail.

pCloud

pCloud was one of the first services I used. It has fast sync speeds and no file size limits, making it ideal for those working with large files. It also has a built-in native media player to view high-resolution images and videos.

pCloud has fast sync speeds and no file size limits, but charges extra for zero-knowledge encryption.

Unfortunately, pCloud charges extra for zero-knowledge encryption, so you’ll pay $4.99 more on top of your standard monthly fee, which is between $4-$8 per month on selected plans. Personal plans max out at 2TB of storage space, which some users may find limiting.


Buy now:


Sync.com

With Sync.com, users don’t need to pay extra for zero-knowledge encryption, as it’s incorporated into all plans, including the free plan. Personal plans also offer up to 6TB of storage space and unlimited monthly data transfers.

Sync.com includes zero-knowledge encryption by default, but sync speeds can be slow.

One limitation: sync speeds are not the quickest, with uploads and download speeds being considerably slower than other options. Also, moving files around can feel cumbersome, with the drag-and-drop function not being the smoothest.


Buy now:


Other zero-knowledge cloud alternatives

If you find these services’ storage space too limited, Mega is a good alternative. Personal plans reach up to 16TB of storage space, and zero-knowledge encryption is included at no extra cost, although monthly plans are costly. Personal plans can cost up to $32 per month, though you can reduce this to $26 if you pay annually.

IDrive is a good option for those not ready to go all-in with cloud storage, as it offers both local and cloud-based backup for your data. You can clone your entire disk (including operating system and applications) and IDrive will mail out a copy in the event of a complete machine malfunction. The cloud aspect works like any other service, allowing you to upload, download and share your files with others, and they’re all protected with zero-knowledge encryption.

The wrap

In closing, my intention here is not to fearmonger. The reality is more brands are using AI imagery. More apps are being made to generate AI imagery, and cloud companies are gaining the power to use our (real) photographs to help it all happen while making more money.

I want no part of that. It’s more important than ever to have complete control over our photographs and to ensure cloud companies don’t maximize their bottom line at the expense of our privacy. To not be complicit in this, we must seek out alternative services to store our images. Not only to ensure our privacy but also to ensure we help protect the future of our industry.



Source link

Continue Reading
Click to comment

Leave a Reply

Camera

Adobe Max Roundup: the demos

Published

on

By

Adobe Max Roundup: the demos


Photo: Mitchell Clark

This year we attended Adobe Max in person, where we got to demo several of the new features in Photoshop, Lightroom, and Adobe Camera Raw. If you missed the announcement, you can read our coverage of it here, though we’ll be covering the hits here.

We documented the demos on our Instagram, but in case you missed it, we’re rounding them up here. We were also able to interview some folks at Adobe while at the show, so stay tuned for more on the future of Photoshop, Lightroom, and other Adobe projects like Content Credentials.

Adobe Camera Raw Adaptive profile

Adobe has added a new profile to Adobe Camera Raw, called Adaptive. It uses AI to analyze what’s in the scene, and adjust exposure, tones, saturation, and other parameters automatically, potentially giving you a better starting point for your own edits. It’s also designed to work with HDR images and produces both HDR and SDR profiles, making it even more useful for those who aren’t used to editing for HDR displays yet.

Since it’s substantially more opinionated than other profiles like Adobe Color or Adobe Landscape, there’s also an amount slider that lets you tone down or turn up the results.

Photoshop automatic distraction removal

Perhaps one of the niftiest features Adobe added to Photoshop is the automatic distraction removal tool. It analyzes your photo for cables, wires, or people that may be in the way of your subject, then automatically fills in the areas they took up.

Lightroom Quick Actions

Lightroom’s new Quick Actions, available on Mobile and Web, will automatically mask parts of your image like subjects, backgrounds, and skies, and let you make adjustments to them.

Lightroom Frame.io integration

Frame.io is now built into Lightroom, letting you access images uploaded to the cloud service. Combine that with Frame.io’s Camera To Cloud feature, available on some Fujifilm, Panasonic, Nikon, Canon, and Leica cameras, and you can take pictures on your camera then watch them appear wirelessly in Lightroom.

Content Credentials

Adobe’s Content Credentials system is part of a larger industry-wide initiative to help prove what content on the web is authentic, and to keep track of what edits have been made to it. While at Max, we got to take a look at the Chrome extension meant to surface the credentials attached to images on social media and other sites, as well as the closed beta site that lets you attach content credentials to your own images, and view what credentials are attached to existing images.

We got to sit down with one of the senior directors of the Content Authenticity Initiative at Adobe while at the conference, so expect a more thorough check-in of the technology to come.

Generative Extend in Premiere Pro

Photoshop has had several generative AI features in beta for a while now, but now Adobe has introduced one for its Premiere video editing software. It lets you extend a clip by up to two seconds, helping you fill in gaps, transitions, or slightly flubbed takes with imagery generated by Adobe’s Firefly model.

According to Meagan Keane, Principal Product Marketing Manager for Adobe Pro Video, the idea came from asking customers what some of their biggest editing pain points were. The Pro Video team was then able to take that to the research team, and the result is Generative Extend.





Source link

Continue Reading

Camera

On this day in 2014: The Panasonic GH4, which brought 4K to the masses, reviewed

Published

on

By

On this day in 2014: The Panasonic GH4, which brought 4K to the masses, reviewed


When you use DPReview links to buy products, the site may earn a commission.
Photo: DPReview staff

As part of DPReview’s 25th anniversary celebration, we’re looking back at some of the more significant cameras to come along over the past 25 years, and today, we’re highlighting the camera that led the 4K revolution in consumer cameras: the Panasonic GH4. Officially announced in early 2014, we published our GH4 review on October 16, 2014 – ten years ago today.

We’ve all become so used to 4K video that it no longer registers as unique. It’s found on virtually every mirrorless camera made today, smartphones, action cameras, miniature gimbal cameras, webcams, security cameras, and even those wacky $100 ‘pro’ video cameras you see on Amazon and wonder who buys them.

But, of course, it wasn’t always that way, and we have the GH4 – the first mirrorless camera to capture 4K video internally – to thank for opening the metaphorical floodgates and ushering in the 4K era.

I had a personal interest in the GH4: as someone who had adopted the GH line early on for video projects, I was as curious as anyone to know how it stacked up. Although I was on staff at DPReview, I wasn’t involved in writing the GH4 review and probably read it with as much anticipation as anyone else.

Photo: DPReview staff

It’s worth remembering that the GH4 was a hybrid camera designed to appeal to both stills and video shooters, and it had features to appeal to those who wanted to do both. Like the GH3, the camera was built around a 16MP sensor, but it supported 12fps burst shooting (7.5fps with focus tracking), a 1/8000 shutter speed and 1/250th flash sync. It even had a rather lovely 2.36M-dot OLED viewfinder.

But realistically, nobody was buying a GH4 just to shoot stills. You bought it because you wanted a solid video camera that could shoot stills when needed.

“Realistically, nobody was buying a GH4 just to shoot stills.”

The GH4 could capture 4K video at up to UHD 4K/30p (3840×2160) or DCI 4K/24p (4096×2160) internally and supported both Long GOP and All-I codecs at bit rates up to 200Mbps. Although we take bit rates like this for granted today, this was very high at the time. The camera also produced pleasing 1080p video, though as we called out in our review, its HD video wasn’t as good as the perfectly oversampled 1080p video from the Sony a7s.

Despite the impressive specs and beautiful video to match, we had some nitpicks. To start, 4K video captured internally only had 8-bit 4:2:0 color, providing less flexibility for color grading in post. (10-bit 4:2:2 color was supported, but only when recording externally.) Additionally, the GH4’s sensor was 4608 pixels wide, requiring the camera to use a smaller, native crop of the sensor when shooting 4K. This resulted in a 1.1x crop for DCI 4K and a 1.2x crop for UHD 4K.

Equally as crucial as its video specs, the GH4 illustrated Panasonic’s commitment to supporting a professional video workflow.

The camera included several now-common tools to improve the video shooting experience, including focus peaking, two zebra settings, control over Master Pedestal (black level) and luminance scale, and a ‘cinema-like’ gamma preset. It also allowed users to set the shutter speed and ISO as shutter angle and gain and could generate color bars for calibration. It was also possible to switch between capture frequencies, meaning the camera could support NTSC, PAL, and true 24fps cinema standards.

“Equally as crucial as its video specs, the GH4 illustrated Panasonic’s commitment to supporting a professional video workflow.”

Alongside the camera, Panasonic released the optional DMW-YAGH interface unit. This $1999 accessory unit added two XLR inputs for audio, an SDI input for timecode, four 3D-SDI connectors capable of outputting 4:2:2 10-bit video and a 12V DC power socket. The unit attached to the bottom of the GH4 and felt oversized relative to the camera. Beginning with the GH5, the interface unit was phased out in favor of the DMW-XLR1, a much more affordable option in the style of the hotshoe-mounted XLR adapter we’ve become accustomed to today.

The GH4 was also the camera Panasonic used to debut its newest autofocus technology: Depth-from-Defocus, or DFD. DFD attempted to build a depth map of a scene by making tiny focus adjustments and analyzing changes in the image. With an understanding of the out-of-focus characteristics of a particular lens, the camera could build a depth map of the scene.

The optional DMW-YAGH ‘Interface Unit’ provided a more extensive selection of video industry connectors for using the GH4 as part of a high-end video rig.

However promising the technology may have been, DFD never quite met expectations. Panasonic really wanted to make it work, and it’s possible that, given fast enough sampling and processing, it might have continued to improve. Unfortunately for Panasonic, cameras using phase-detect autofocus consistently provided a better AF experience, particularly when shooting video, and the company eventually made the jump to phase-detection with the Lumix S5II in 2023, finally arriving in the GH series on the GH7 in 2024.

In our review of the GH4, we found a lot to like and a few frustrations. For example, despite having an autoexposure compensation dial, the camera didn’t allow you to use it when using Auto ISO in manual exposure mode, and there was no Auto ISO option when shooting video in M mode. Overall, though, we were mighty impressed and saved our biggest praise for the camera’s video capabilities:

“The GH4 was also the camera Panasonic used to debut its newest autofocus technology: Depth-from-Defocus, or DFD.”

“It’s in terms of video that the GH4 really stands out. It produces some of the best video we’ve yet seen – losing out only to the Sony a7S’s moiré-free 1080 output. The ability to capture good quality 4K, whether for use at full resolution, downsampling to 1080 or cropping to 1080, adds real flexibility to the camera. Low light performance is solid if not exceptional,” we concluded.

The GH4 landed in the retail market at a price of $1699, or about $2260 today adjusted for inflation, which isn’t far off the $2199 price of the GH7. It’s amazing to think about how expectations for video have changed over the years. However, the GH7 has its work cut out for it: rather than being an obvious standout in the crowd, it has to compete in a marketplace of cameras brimming with video features. Maybe in another ten years, we’ll look back to see how it held up.



Source link

Continue Reading

Camera

DJI's new dual-camera Air 3S drone gets a larger sensor and LiDAR

Published

on

By

DJI's new dual-camera Air 3S drone gets a larger sensor and LiDAR


Photo: DJI

DJI has announced the Air 3S, an update to its existing dual-camera Air 3 model that upgrades the primary camera to a larger Type 1 sensor and adds features that promise to make the drone easier to operate at night. DJI is positioning the Air 3S as “perfect for travel photography.”

The main camera on the Air 3S gains a 50MP Type 1 (13.2 x 8.8mm) CMOS sensor and features a 24mm equiv. F1.8 lens. That’s an upgrade from its predecessor, which utilized a smaller Type 1/1.3 (4.8 x 3.6mm) CMOS sensor with a marginally faster F1.7 lens. The telephoto camera remains unchanged between the two models, using a 48MP Type 1/1.3 sensor and featuring a 70mm equiv. F2.8 lens.

In addition to stills, both cameras can capture up to 4K/120p or 4K/60p when shooting HDR. All video can be captured in 10-bit, even in regular color mode, and D-Log M and HLG modes are available. The maximum ISO has been raised to 12,800 (3200 when shooting D-Log M). However, DJI says the Air 3S includes a new, more advanced video encoding algorithm that reduces video file size by over 30% without compromising image quality.

DJI claims both cameras can capture up to 14 stops of dynamic range and says the new main camera should capture even more detail than the camera on its more expensive Mavic 3 Pro model.

The DJI Air 3S is similar to its predecessor, the Air 3, but it gains a larger Type 1 (13.2 x 8.8mm) CMOS sensor on its main camera.

Image: DJI

Beyond the cameras, DJI has added several appealing features to the Air 3S:

A new Free Panorama mode enables users to create panoramic shots by stitching together images across a manually selected subject area. This works with either camera, but DJI suggests the best results will come from using the telephoto camera, which reduces distortion.

The Air 3S also includes features designed to make it easier and safer to fly at night, including Nightscape Obstacle Sensing, with the Air 3S becoming the first DJI drone to feature forward-facing LiDAR. Additionally, the drone includes downward-facing infrared sensors and six vision sensors (two each at the front, rear and bottom). According to DJI, this combination of sensors provides the Air 3S with “nightscape omnidirectional obstacle sensing,” which should allow the drone to automatically identify and navigate around obstacles for safer nighttime photography.

This technology also enables DJI’s next-gen Smart RTH (return-to-home) feature, intended to allow the drone to return safely to its takeoff location, even at night.

The Air 3S is available with either DJI’s RC-N3 controller, which requires a smartphone to monitor the camera feed and aircraft status, or the RC 2 controller (above), which includes a built-in 700-nit 5.5″ 1080 screen.

Image: DJI

The Air 3S also includes real-time vision positioning and map construction technology, designed to allow the drone to memorize a flight path and to return safely when adequate light is available, even in areas without satellite coverage.

The new model also features DJI’s ActiveTrack 360 subject tracking but introduces a new subject focusing feature designed to keep a subject in sharp focus, even during manual flight or when a subject moves off-center. This should allow a pilot to focus on creative decisions like composition or camera movement while ActiveTrack keeps the subject in focus.

There are a few hardware specs worth noting as well. The Air 3S weighs 724g (1.6 lbs), just 4g more than its predecessor, and is rated for 45 minutes of flight time. It includes DJI’s O4 video transmission system that transmits 10-bit video at up to 1080/60p and 42GB of built-in storage. A new Off-state Quick Transfer feature allows files to be transferred from the drone to a smartphone or a computer even when powered off.

Finally, for the privacy-conscious, a new Local Data mode completely disconnects the drone from the internet, ensuring that all data stays only on the device. DJI likens this mode to airplane mode on a smartphone.

The Air 3S is available in several packages. The Fly More combo shown above includes the RC 2 controller, ND filter set, two additional batteries (for a total of three), a charging hub, extra props and a shoulder bag, will retail for $1599.

Image: DJI

The charging hub that ships with the Air 3S supports PD fast charging and features a power accumulation function. It allows users to transfer the remaining power from several depleted batteries into the battery with the most remaining power – something anyone who has had to use drones in remote locations without a charging station is likely to appreciate.

Price and availability

The DJI Air 3S is available for purchase in several configurations: the drone with the RC-N3 controller, which requires a smartphone to monitor the camera feed and flight status, will retail for $1099. A Fly More combo with the RC-N3 controller, ND filter set, two additional batteries, a battery charging hub and shoulder back will retail for $1399. Finally, a Fly More combo with DJI’s RC 2 controller, which includes a built-in 700-nit 5.5″ 1080p screen, ND filter set, two additional batteries, charging hub and shoulder pack, will retail for $1599.



Source link

Continue Reading

Trending