Bone-Shaking Haunted Mirror Uses Stable Diffusion

We once thought that the best houses on Halloween were the ones that gave out full-size candy bars. While that’s still true, these days we’d rather see a cool display of some kind on the porch. Although some might consider this a trick, gaze into [Tim]’s mirror and you’ll be treated to a spooky version of yourself.

Here’s how it works: At the heart of this build is a webcam, OpenCV, and a computer that’s running the Stable Diffusion AI image generator. The image is shown on a monitor that sits behind 2-way mirrored glass.

We really like the frame that [Tim] built for this. Unable to find something both suitable and affordable, they built one out of wood molding and aged it appropriately.

We also like the ping pong ball vanity globe lights and the lighting effect itself. Not only is it spooky, it lets the viewer know that something is happening in the background. All the code and the schematic are available if you’d like to give this a go.

There are many takes on the spooky mirror out there. Here’s one that uses a terrifying 3D print.

WhisperFrame Depicts The Art Of Conversation

At this point, you gotta figure that you’re at least being listened to almost everywhere you go, whether it be a home assistant or your very own phone. So why not roll with the punches and turn lemons into something like a still life of lemons that’s a bit wonky? What we mean is, why not take our conversations and use AI to turn them into art? That’s the idea behind this next-generation digital photo frame created by [TheMorehavoc].
Essentially, it uses a Raspberry Pi and a Respeaker four-mic array to listen to conversations in the room. It listens and records 15-20 seconds of audio, and sends that to the OpenWhisper API to generate a transcript.
This repeats until five minutes of audio is collected, then the entire transcript is sent through GPT-4 to extract an image prompt from a single topic in the conversation. Then, that prompt is shipped off to Stable Diffusion to get an image to be displayed on the screen. As you can imagine, the images generated run the gamut from really weird to really awesome.

The natural lulls in conversation presented a bit of a problem in that the transcription was still generating during silences, presumably because of ambient noise. The answer was in voice activity detection software that gives a probability that a voice is present.

Naturally, people were curious about the prompts for the images, so [TheMorehavoc] made a little gallery sign with a MagTag that uses Adafruit.io as the MQTT broker. Build video is up after the break, and you can check out the images here (warning, some are NSFW).

Continue reading “WhisperFrame Depicts The Art Of Conversation”

E-Paper News Feed Illustrates The Headlines With AI-Generated Images

It’s hard to read the headlines today without feeling like the world couldn’t possibly get much worse. And then tomorrow rolls around, and a fresh set of headlines puts the lie to that thought. On a macro level, there’s not much that you can do about that, but on a personal level, illustrating your news feed with mostly wrong, AI-generated images might take the edge off things a little.

Let us explain. [Roy van der Veen] liked the idea of an e-paper display newsfeed, but the crushing weight of the headlines was a little too much to bear. To lighten things up, he decided to employ Stable Diffusion to illustrate his feed, displaying both the headline and a generated image on a 7.3″ Inky 7-color e-paper display. Every five hours, a script running on a Raspberry Pi Zero 2W fetches a headline from a random source — we’re pleased the list includes Hackaday — and composes a prompt for Stable Diffusion based on the headline, adding on a randomly selected prefix and suffix to spice things up. For example, a prompt might look like, “Gothic painting of (Driving a Motor with an Audio Amp Chip). Gloomy, dramatic, stunning, dreamy.” You can imagine the results.

We have to say, from the examples [Roy] shows, the idea pretty much works — sometimes the images are so far off the mark that just figuring out how Stable Diffusion came up with them is enough to soften the blow. We’d have preferred if the news of the floods in Libya had been buffered by a slightly less dismal scene, but finding out that what was thought to be a “ritual mass murder” was really only a yoga class was certainly heartening.

Turning Old Kindles Into AI Powered Picture Frames

While we tend to think of Amazon’s e-paper Kindles as more or less single-purpose devices (which to be fair, is how they’re advertised), there’s actually a full-featured Linux computer running behind that simple interface, just waiting to be put to work. Given how cheap you can get old Kindles on the second hand market, this has always struck us as something of a wasted opportunity.

This is why we love to see projects like Kindlefusion from [Diggedypomme]. It turns the Kindle into a picture frame to show off the latest in machine learning art thanks to Stable Diffusion. Just connect your browser to the web-based control interface running on the Kindle, give it a prompt, and away it goes. There are also functions to recall previously generated images, and if you’re connecting from a mobile device, support for creating images from voice prompts.

You can find cheap older Kindles on eBay.

All you need is a Kindle that can be jailbroken, though technically the software has only been tested against older third and fourth-generation hardware. From there you install a few required packages as listed in the project documentation, including Python 3. Then you just move the Kindlefusion package over either via USB or SSH, and do a little final housekeeping before starting it up and letting it take over the Kindle’s normal UI.

Given the somewhat niche nature of Kindle hacking, we’re particularly glad to see that [Diggedypomme] went through the trouble of explaining the nuances of getting the e-reader ready to run your own code. While it’s not difficult to do, there are plenty of pitfalls if you’ve never done it before, so a concise guide is a nice thing to have. Unfortunately, it seems like Amazon has recently gone on the offensive, with firmware updates blocking the exploits the community was using for jailbreaking on all but the older models that are no longer officially supported.

While it’s a shame you can’t just pick up a new Kindle and start hacking (at least, for now), there are still millions of older devices floating around that could be put to good use. Hopefully, projects like this can help inspire others to pick one up and start experimenting with what’s possible.

Continue reading “Turning Old Kindles Into AI Powered Picture Frames”

Getty Images Is Suing An AI Image Generator For Using Its Images

As per the Getty Images legal complaint, the Stable Diffusion AI seems to reproduce gooey versions of the Getty Images watermark in some of its output. Credit: Getty Images

Many AI systems require huge training datasets in order to achieve their impressive feats. This applies whether or not you’re talking about an AI that works with images, natural language, or just about anything else. AI developers are starting to come under scrutiny for where they’re sourcing their datasets. Unsurprisingly, stock photo site Getty Images is at the forefront of this, and is now suing the creators of Stable Diffusion over the matter, as reported by The Verge.

Stability AI, the company behind Stable Diffusion, is the target of the lawsuit for one good reason: there’s compelling evidence the company used Getty Images content without permission. The Stable Diffusion AI has been seen to generate output images that actually include blurry approximations of the Getty Images watermark. This is somewhat of a smoking gun to suggest that Stability AI may have scraped Getty Images content for use as training material.

The copyright implications are unclear, but using any imagery from a stock photo database without permission is always asking for trouble. Various arguments will likely play out in court. Stability AI may make claims that their activity falls under fair use guidelines, while Getty Images may claim that the appearance of perverted versions of their watermark may break trademark rules. The lawsuit could have serious implications for AI image generators worldwide, and is sure to be watched closely by the nascent AI industry. As with any legal matter, just don’t expect a quick answer from the courts.

[Thanks to Dan for the tip!]

3D Printed Triptych Shows Trio Of AI-Generated Images

Fascinated by art generated by deep learning systems such as DALL-E and Stable Diffusion? Then perhaps a wall installation like this phenomenal e-paper Triptych created by [Zach Archer] is in your future.

The three interlocking frames were printed out of “Walnut Wood” HTPLA from ProtoPasta, and hold a pair of 5.79 inch red/black/white displays along with a single 7.3 inch red/yellow/black/white panel from Waveshare. There are e-paper panels out there with more colors available if you wanted to go that route, but judging by the striking images [Zach] has posted, the relatively limited color palettes available on these displays doesn’t seem to be a hindrance.

Note the clever S-shaped brackets holding in the displays.

To create the images themselves, [Zach] wrote a script that would generate endless customized portraits using Stable Diffusion v1.4, and then manually selected the best to get copied over to a 32 GB micro SD card. The side images were generated on the dreamstudio.ai website, and also dumped on the card.

Every 12 hours a TinyPico ESP32 development board in the frame picks some images from the card, applies the necessary dithering and color adjustments to make them look good on the e-paper, and then updates the displays. Continue reading “3D Printed Triptych Shows Trio Of AI-Generated Images”

Giving Stable Diffusion Some Depth

You’ve likely heard quite a bit of buzz over the last few months about Stable Diffusion. The new version (v2) has come out, and in addition to the standard image-to-image and text-to-image modes, it also has a depth-image-to-image that can be incredibly useful. [Andrew] has a write-up that guides you on using this mode.

The basic idea is that you can take both an image and depth into the model, which allows you to control what gets put where. Stable Diffusion is a bit confusing, but we already have some great resources to wrap your head around it. In terms of input, you can use a depth map from a camera with lidar (many recent phones include this) or have another model (like MiDaS) estimate it from a 2D picture. This becomes powerful when you can preserve a specific composition, such as an iconic scene from a well-known movie. You can keep the characters’ poses on the screen but transform the style of the scene into whatever you wish (as seen above).

We have already covered a technique to generate textures right in blender, but this new depth information has already been implemented to provide better accuracy of the textures.

[Justin Alvey] used it to create architectural photos from dollhouse furniture. Using the MiDaS model, he estimated the depth and threw away the RGB aspects by setting the denoising strength to maximum. The simplified dollhouse furniture was easily recognizable to the model, which helped produce great results.

However, the only downside is that the perspective produces a rather dollhouse feel. Changing the focal length and moving farther away helps. Overall, it’s a clever use of what the new AI model can do. It’s a fast-moving space, so this will likely be out of date in a few months.