Vision


Simple Face Recognition: New in Wolfram Language 12

#artificialintelligence

A face recognition network can be trained in such a way that the Euclidean distance in the embedding feature space directly corresponds to face similarity. Using the embedding as facial descriptors, we can implement a simple face recognition algorithm without the need to train a new model. Here is a small sample of a set of images from a family of five. You can use FeatureSpacePlot to visualize the descriptors in clusters. Use Classify to perform classification in the feature space.


New study shows how friendlier facial expressions may have helped humans evolve

Daily Mail - Science & tech

A new study suggest that ability to convey kindness through facial expressions may have been a key factor in human evolution. The study was conducted by Matteo Zanella and a team of researchers at the University of Milan, and published this week in Science Advances. The team compared genetic data from human stem cells with samples from the remains of two Neanderthals and one Denisovan, a sister species to Neanderthals found in central Asia. They specifically focused on the BAZ1B gene, which has been connected to Williams-Beuren syndrome, a condition that causes people to develop wide mouths and small noses that give a generally kind and welcoming impression. The BAZ1B gene has also been associated with the evolution of two extra muscles in dogs that allow them to widen and narrow their eyes in expressive ways, something wolves aren't able to do.


Big Brother is watching: Chinese city with 2.6m cameras is world's most heavily surveilled

#artificialintelligence

Qiu Rui, a policeman in Chongqing, was on duty this summer when he received an alert from a facial recognition system at a local square. There was a high probability a man caught on camera was a suspect in a 2002 murder case, the system told him. The depth, breadth and intrusiveness of China's mass surveillance may be unprecedented in modern history The city's surveillance system scans facial features of people on the streets from frames of video footage in real time, creating a virtual map of the face. It can then match this information against scanned faces of suspects in a police database. If there is a match that passes a preset threshold, typically 60% or higher, the system immediately notifies officers.


Big Brother is watching: Chinese city with 2.6m cameras is world's most heavily surveilled

#artificialintelligence

Qiu Rui, a policeman in Chongqing, was on duty this summer when he received an alert from a facial recognition system at a local square. There was a high probability a man caught on camera was a suspect in a 2002 murder case, the system told him. The depth, breadth and intrusiveness of China's mass surveillance may be unprecedented in modern history The city's surveillance system scans facial features of people on the streets from frames of video footage in real time, creating a virtual map of the face. It can then match this information against scanned faces of suspects in a police database. If there is a match that passes a preset threshold, typically 60% or higher, the system immediately notifies officers.


Build your First Multi-Label Image Classification Model in Python

#artificialintelligence

Are you working with image data? This got me thinking -- what can we do if there are multiple object categories in an image? Making an image classification model was a good start, but I wanted to expand my horizons to take on a more challenging task -- building a multi-label image classification model! I didn't want to use toy datasets to build my model -- that is too generic. And then it struck me -- movie/TV series posters contain a variety of people. Could I build my own multi-label image classification model to predict the different genres just by looking at the poster?


A 3D Printer Powered by AI and Machine Vision

#artificialintelligence

Objects made with 3D printing can be lighter, stronger, and more complex than those produced through traditional manufacturing methods. But several technical challenges must be overcome before 3D printing transforms the production of most devices. Commercially available printers generally offer only high-speed, high-precision, or high-quality materials. Rarely do they offer all three, limiting their usefulness as a manufacturing tool. Today, 3D printing is used mainly for prototyping and low-volume production of specialized parts.


Lego Finds An Inventive Way to Combine AI and Motion Tracking

#artificialintelligence

Lego toy systems have been around for generations and have been considered by many as a way to stimulate the imagination. Quite a few users have at some point imagined having a Lego figure in their own image they could use with their sets. Realizing that fact, Lego has decided to try and make that dream come true. As Gizmodo reports, Lego will try to realize that dream for anybody who visits there theme park that will open in New York in 2020. To do this the company will employ sophisticated motion tracking and neural network facial recognition.


Spotting drivers on their phone is just the tip of the iceberg for AI-enabled cameras

#artificialintelligence

Last week, the Australian state of New South Wales announced a plan to crack down on drivers using their phones on the road. The state's transport agency said it had integrated machine vision into roadside cameras to spot offenders. The AI automatically flags suspects, humans confirm what's going on, and a warning letter is sent out to the driver. "It's a system to change the culture," the assistant police commissioner of New South Wales, Michael Corboy, told Australian media, noting that police hoped the technology would cut fatalities on the road by a third over two years. It seems an admirable scheme, top to bottom.


A Surrogate Video-Based Safety Methodology for Diagnosis and Evaluation of Low-Cost Pedestrian-Safety Countermeasures: The Case of Cochabamba, Bolivia

#artificialintelligence

Due to a lack of reliable data collection systems, traffic fatalities and injuries are often under-reported in developing countries. Recent developments in surrogate road safety methods and video analytics tools offer an alternative approach that can be both lower cost and more time efficient when crash data is incomplete or missing. However, very few studies investigating pedestrian road safety in developing countries using these approaches exist. This research uses an automated video analytics tool to develop and analyze surrogate traffic safety measures and to evaluate the effectiveness of temporary low-cost countermeasures at selected pedestrian crossings at risky intersections in the city of Cochabamba, Bolivia. Specialized computer vision software is used to process hundreds of hours of video data and generate data on road users' speed and trajectories.


China's top fintech adviser sees 'bright prospects' for facial recognition payment systems

#artificialintelligence

China's top tech adviser for the financial sector sees "bright prospects" for facial recognition payment systems despite lingering concerns over privacy and data security. The artificial recognition-powered payment method is expected to be widely used across the country because it offers convenience and efficiency, according to Chen Jing, a member of the State Council's Advisory Committee for State Information and former chief of the central bank's technology division. Still, regulations need to be in line with the technology, which means limiting use to qualified service providers and capping the amount of payment allowed, Chen told The Beijing News in a report on Thursday. "Development is of overriding importance because falling behind is the biggest cause for concern," Chen was quoted as saying. "Through dedicated exploration and efforts, face-scan payment will win recognition from society."