AI Camera Knows Its S**t

[Caleb] shares a problem with most dog owners. Dogs leave their… byproducts…all over your yard. Some people pick it up right away and some just leave it. But what if your dog has run of the yard? How do you know where these piles are hiding? A security camera and AI image detection is the answer, but probably not the way that you think.

You might think as we did that you could train the system to recognize the–um–piles. But instead, [Caleb] elected to have the AI do animal pose estimation to detect the dog’s posture while producing the target. This is probably easier than recognizing a nondescript pile and then it doesn’t matter if it is, say, covered with snow.

Continue reading “AI Camera Knows Its S**t”

OpenCV Knows Where Your Hand Is

We have to say, [Murtaza]’s example game in his latest video isn’t very exciting. However, the OpenCV technique he uses to track a hand and determine its distance from a single camera is pretty interesting. The demo shows a random button on the screen and you have to use your hand to press the button which then moves so you can try again. The hand measurement seems accurate to a few centimeters which is good enough for many applications.

The Python code is actually quite straightforward. Essentially, the software tracks your hand and by estimating its relative size to determine how far away it is. Of course, your hand might also rotate, and [Murtaza] works through all the cases step-by-step. If we wanted to know a distance, we’d probably turn to ultrasonics or a time of flight sensor. The problem is, those sensors can’t tell your hand from anything else that happens to be in front of it. The use of a single camera to track and locate is pretty impressive.

If you haven’t used OpenCV before, the channel has a lot of tutorials and they are all worth watching. Computer vision is a great technique and can replace a lot of things in some applications. GPS, for example. Or, try this creepier tracking application next Halloween.

Continue reading “OpenCV Knows Where Your Hand Is”

The Seductive Pull Of An Obsolete Home Movie Format

It’s dangerous for a hardware hacker to go into a second-hand store. I was looking for a bed frame for my new apartment, but of course I spent an age browsing all the other rubbish treasures on offer. I have a rough rule of thumb: if it’s not under a tenner and fits in one hand, then it has to be exceptional for me to buy it, so I passed up on a nice Grundig reel-to-reel from the 1960s and instead came away with a folding Palm Pilot keyboard and a Fuji 8mm home movie camera after I’d arranged delivery for the bed. On those two I’d spent little more than a fiver, so I’m good. The keyboard is a serial device that’s a project for a rainy day, but the camera is something else. I’ve been keeping an eye out for one to use for a Raspberry Pi camera conversion, and this one seemed ideal. But once I examined it more closely, I was drawn into an unexpected train of research that shed some light on what must of been real objects of desire for my parents generation.

A Thrift Store Find Opens A Whole New Field

One f the surprises comes in just how small this thing is.
One of the surprises comes in just how small this thing is.

The Fuji P300 from 1972 is typical among consumer movie cameras of the day. It takes the form of a film magazine with a zoom lens assembly on its front, a reflex viewfinder on its side, and a handle with a shutter trigger button on it protruding vertically below the magazine and also housing the batteries.

Surprisingly it still has a mercury cell that would have powered its light meter; a minor annoyance to dispose of this correctly. Sometimes these devices had clockwork motors, but this one has an electric motor. It also has a light sensor that is coupled to some kind of electromechanical aperture. It would have been an expensive camera when it was new, probably as much of a purchase as an SLR or a decent mirrorless camera here in 2021.

The surprise came when I opened it up, for it looked like no other 8mm camera I had seen. I’m familiar wit the two reels of a Standard 8 or the boxy cassette of Super 8, but this one used something different. That film magazine is made to fit a compact twin-reel cartridge whose film fits in a metal film gate. This is a Single 8 camera, Fuji’s entry in the all-in-one 8 mm film market, and a format I never knew existed. To explain my unexpected discovery it was necessary to delve into the world of home movie formats in the decade before videotape arrived and drove them out. Continue reading “The Seductive Pull Of An Obsolete Home Movie Format”

The insides of an oscilloscope

Oscilloscope Probes Itself To Add Video

Modern oscilloscopes are often loaded with features, but every now and then you run into a feature that seems easy to implement yet isn’t available. [kgsws] wanted to use his Rigol DS1074 to show live measurements in his YouTube videos, but found out that this scope doesn’t support video output. Not to be deterred, [kgsws] decided to add this feature himself. In the video embedded below, he describes in detail the process of adding a USB Video Capture (UVC) interface to his oscilloscope.

The basic idea was to find the signals going into the scope’s display and read them out using a Cypress EZ-USB board. This is a development board that can be used to design USB devices, and supports the UVC mode. However, with no documentation of any of the Rigol’s internal circuitry [kgsws] had to probe the display connector to find out which pin carried which signal. And since he had no other scope available than this Rigol, he hooked up the various bits of the disassembled instrument so that it could (awkwardly) probe its own internal signals.

After mapping out its own display signals, it was time to hook them up to the EZ-USB board. [kgsws] achieved this by soldering about two dozen tiny wires to SMD pads on the motherboard. The EZ-USB board itself was placed in the back of the scope’s case, but had to be stripped of unneeded components in order to save space and power. A very clever trick was the addition of a reed switch, which allowed [kgsws] to set the EZ-USB board to programming mode without having to open the scope’s case, by simply holding a magnet near the switch.

After soldering a USB connector into a spare slot in the RF shield the project was complete. The Rigol can now be connected to a PC and will simply appear as a video capture device, ready to be streamed or captured for [kgsws]’s future project videos. We’ve seen other hacks on the Rigol DS1000Z series to capture a series of screenshots or to enable additional bandwidth and features, but adding a live video output was not one of the options so far. Continue reading “Oscilloscope Probes Itself To Add Video”

ESP32 Clock Pushes Outrun Graphics Over Composite

We’ve covered plenty of clocks powered by the ESP32, but this one from [Marcio Teixeira] is really something special. Rather than driving a traditional physical display, the microcontroller is instead generating a composite video signal of an animated digital clock. This could be fed into whatever device you wish, but given the 80’s synthwave style it’s pumping out, you’ll probably want to find a suitably retro CRT to do it justice.

Specifically this is a variant of the “Dali” clock, where each digit seems to melt and morph into its successor. Though his version doesn’t necessarily share code with all the previous iterations, [Marcio] does credit the developers who have pulled off similar visual tricks going all the way back to 1979. Given the vintage of this particular animation, the neon skyline and infinite scrolling grid certainly feel like a perfect fit.

Want to add a little vaporwave vibe to your own workbench? Assuming you’ve already got a 80s style CRT, all you need is an ESP32 and two wires stuck into the composite video port. One goes to ground, and the other goes to the chip’s analog pin. Once everything is powered up, you’ll be able to configure the clock with a web-based interface. It doesn’t get much easier than that.

In the documentation, [Marcio] calls out a few open source projects which were instrumental to getting his clock off the ground. The pioneering work [bitluni] did to get video out of the ESP32 is something of a given, but he also sends a hat tip to [rossumur] for his collection of 8-bit game console emulators written for the microcontroller. Projects like this are a fantastic example of what’s possible when a community works together to truly push the envelope.

Continue reading “ESP32 Clock Pushes Outrun Graphics Over Composite”

Super 8 Camera Brought Back To Life

The Super 8 camera, while a groundbreaking video recorder in its time, is borderline unusable now. Even if you can get film for it (and afford its often enormous price), it still only records on 8mm film which isn’t exactly the best quality of film around, not to mention that a good percentage of these cameras couldn’t even record audio. They were largely made obsolete by camcorders in the late ’80s and early ’90s, although some are still used for niche artistic purposes. If you’d rather not foot the bill for the film, though, you can still put one of these to work with the help of a Raspberry Pi.

[befinitiv] has a knack for repurposing antique analog equipment like this while preserving its aesthetic. While the bulk of the space inside of this camera would normally be used for housing film, this makes a perfect spot to place a Raspberry Pi Zero, a rechargeable battery, and a power converter circuit all in a 3D printed enclosure that snaps into the camera just as a film roll would have. It uses the Pi camera module but still makes use of the camera’s built in optics which include a zoom function. [befinitiv] also incorporated the original record button so that from the outside this looks like a completely unmodified Super 8 camera.

The camera can connect to a WiFi network and can stream live video to a computer, or it can record video files to an internal SD card. As a bonus, thanks to the power converter circuit, it is also capable of charging a cell phone. [befinitiv] notes that many of the aesthetic properties of 8 mm film seem to be preserved when using this method, and he has several theories as to why but no definitive answer. If you’d like to take a look at some of his other projects like this, check out this analog camera that is now able to take digital pictures. Continue reading “Super 8 Camera Brought Back To Life”

Mastering Stop Motion Through Machine Learning

Stop motion animation is notoriously difficult to pull off well, in large part because it’s a mind-numbingly slow process. Each frame in the final video is a separate photograph, and for each one of those, the characters and props need to be moved the appropriate amount so that the final result looks smooth. You don’t even want to know how long Ben Wyatt spent working on Requiem for a Tuesday, though to be fair, it might still get done before the next Avatar.

But [Nick Bild] thinks his latest project might be able to improve on the classic technique with a dash of artificial intelligence provided by a Jetson Xavier NX. Basically, the Jetson watches the live feed from the camera, and using a hand pose detection model, waits until there’s no human hand in the frame. Once the coast is clear, it takes a shot and then goes back to waiting for the next hands-free opportunity. With the photographs being taken automatically, you’re free to focus on getting your characters moving around in a convincing way.

If it’s still not clicking for you, check out the video below. [Nick] first shows the raw unedited video, which primarily consists of him moving three LEGO figures around, and then the final product produced by his system. All the images of him fiddling with the scene have been automatically trimmed, leaving behind a short animated clip of the characters moving on their own.

Now don’t be fooled, it’s still going to take awhile. By our count, it took two solid minutes of moving around Minifigs to produce just a few seconds of animation. So while we can say its a quicker pace than with traditional stop motion production, it certainly isn’t fast.

Machine learning isn’t the only modern technology that can simplify stop motion production. We’ve seen a few examples of using 3D printed objects instead of manually-adjusted figures. It still takes a long time to print, and of course it eats up a ton of filament, but the mechanical precision of the printed scenes makes for a very clean final result.

Continue reading “Mastering Stop Motion Through Machine Learning”