DensePose From WiFi - THE A.I. Dilema

I was working in new media art and using Sony X Box's dot plotter for gesture based interactions going as far back at 2008, the licensed version of the gesture tech was super expensive and not for the kind of experiments I was up to. There were dozens of us globally, collaboratively trying this hacking for fun & artistic reasons.

MAPPING WIFI SIGNALS IN 3 DIMENSIONS (2015)

[Charles] is on a quest to complete?ever more jaw-dropping hacks with the popular low-cost ESP8266 WiFi modules. This week’s project is?plotting WiFi received signal strength in 3D space . While the ESP8266 is capable of providing a Received Signal Strength Indication (RSSI), [Charles] didn’t directly use it. He wrote a simple C program on his laptop to?ping the ESP8266 at around 500Hz. The laptop would then translate the RSSI from the ping replies to a color value, which it would then send to the ESP8266. Since the ESP8266 was running [Charles’] custom firmware (as seen in his?WiFi cup project ), it could directly display the color on a WS2812 RGB LED.

The colors seemed random at first, but [Charles] noticed that there was a pattern. He just needed a way to visualize the LED over time. A single frame long exposure would work, but so would video. [Charles] went the video route, creating?SuperLongExposure , an FFMPEG-based tool which extracts every video frame and composites them into a single frame. What he saw was pretty cool – there were definite stripes of good and bad signal.

Armed with this information, [Charles] went for broke and mounted his ESP8266 on a large gantry style mill. He took several long exposure videos of a 360x360x180mm area. The videos were extracted into layers. The whole data set could then be visualized with Voxeltastic, [Charles’] own HTML5/WEBGL based render engine. The results were?nothing short of amazing. ?The signal strength increases and decreases in nodes and anti-nodes which correspond to the 12.4 cm wavelength of a WiFi signal. The final render looks incredibly organic, which isn’t completely surprising. We’ve seen the same kind of image from commercial antenna simulation characterization systems.

Once again [Charles] has blown us away, we can’t wait to see what he does next!

Response from a community member to this project: Truth?says: on February 17, 2015 at 11:00 am One problem would be that moving a large bag of water and minerals through any RF field will change it. Like most things technical, if you could remove the users from the system it would be so much easier.

Watching more behind these projects reminds me of what was harmless and fun, has taken a whole new meaning. Good or bad time will tell. I see it used in disaster management already.

dvances in computer vision and machine learning techniques have led to significant development in 2D and 3D human pose estimation from RGB cameras, LiDAR, and radars. However, human pose estimation from images is adversely affected by occlusion and lighting, which are common in many scenarios of interest. Radar and LiDAR technologies, on the other hand, need specialized hardware that is expensive and power-intensive. Furthermore, placing these sensors in non-public areas raises significant privacy concerns. To address these limitations, recent research has explored the use of WiFi antennas (1D sensors) for body segmentation and key-point body detection. This paper further expands on the use of the WiFi signal in combination with deep learning architectures, commonly used in computer vision, to estimate dense human pose correspondence. We developed a deep neural network that maps the phase and amplitude of WiFi signals to UV coordinates within 24 human regions. The results of the study reveal that our model can estimate the dense pose of multiple subjects, with comparable performance to image-based approaches, by utilizing WiFi signals as the only input. This paves the way for low-cost, broadly accessible, and privacy-preserving algorithms for human sensing.

Research source: https://arxiv.org/abs/2301.00250

Human Activity Detection Using WiFi Signals and Deep Networks

A while ago https://twitter.com/G4lile0 developed Heimdall-WiFi-Radar, only with 3× ESP8266 it was possible to track and position WiFis devices through walls, now with the help of AI we have a new level, it is possible to know where you are and what are you doing!

Source: https://twitter.com/G4lile0/status/1614553700067835906/photo/1

The source code is here https://github.com/G4lile0/Heimdall-WiFi-Radar (Handle with care).