NextMind’s BCI

NextMind, a Paris-based brain-computer interface (BCI) startup, debuted a $400 neural interface dev kit. NextMind’s device is a non-invasive electroencephalogram (EEG), which is a well established method of measuring the voltage fluctuations of neurons from outside the skull. EEGs have been used in medicine, neurology, cognitive science, and a number of related fields too, although unlike more invasive methods out there you might describe it as metaphorically trying to figure out what’s happening in a stadium by listening to the crowd’s roar. 

Attaching to the back of the head with a simple forehead strap, eight prong-like electrodes pick up brain waves from the visual cortex. The wireless device communicates via Bluetooth, and does a portion of the processing on-device whilst offloading the machine learning tasks to the same PC driving the VR experience. The company is pitching a number of use cases, one of which was its potential application in VR headsets. It’s clear that the next generations of VR headsets are heading down the path of integrated eye-tracking though, EEG data being able to work in concert.

More information:

https://www.roadtovr.com/ces-2020-nextmind-400-brain-computer-interface-developer-kit/

HDR Panasonic’s VR Glasses

The micro OLED panels, co-developed by Panasonic and Kopin, are extremely high resolution with almost no hint of the screen-door effect that plagues most VR hardware. They are also the first VR glasses to support HDR with light realistically bouncing off golden decorations. 

Panasonic has made use of its own audio technology in the headset, with Technics drivers in the earbuds providing rich, dynamic sound. The company says it also used optical designs from the Lumix camera division and similar signal processing technologies as found in its TVs and Blu-ray players.

More information:

https://www.theverge.com/platform/amp/2020/1/9/21058056/panasonic-vr-glasses-hdr-hands-on-ces-2020

Deep Learning Model for EEG-based Emotion Recognition

Recent advances in machine learning have enabled the development of techniques to detect and recognize human emotions. Some of these techniques work by analyzing electroencephalography (EEG) signals, which are essentially recordings of the electrical activity of the brain collected from a person's scalp. Most EEG-based emotion classification methods introduced over the past decade or so employ traditional machine learning (ML) techniques such as support vector machine (SVM) models, as these models require fewer training samples and there is still a lack of large-scale EEG datasets. Recently, however, researchers have compiled and released several new datasets containing EEG brain recordings. The release of new datasets opens exciting new possibilities for EEG-based emotion recognition, as they could be used to train deep-learning models that achieve better performance than traditional ML techniques. Unfortunately, however, the low resolution of EEG signals contained in these datasets could make training deep-learning models rather difficult. 

To enhance the resolution of available EEG data, researchers first generated so-called topology-preserving differential entropy features using the electrode coordinates at the time when the data was collected. Subsequently, they developed a convolutional neural network (CNN) and trained it on the updated data, teaching it to estimate three general classes of emotions (i.e., positive, neutral and negative). The researchers trained and evaluated their approach on the SEED dataset, which contains 62-channel EEG signals. They found that their method could classify emotions with a remarkable average accuracy of 90.41 percent, outperforming other machine-learning techniques for EEG-based emotion recognition. In the future, the method proposed by researchers could inform the development of new EEG-based emotion recognition tools, as it introduces a viable solution for overcoming the issues associated with the low-resolution of EEG data. The same approach could also be applied to other deep-learning models for the analysis of EEG data.

More information:

https://techxplore.com/news/2019-12-deep-eeg-based-emotion-recognition.html

Racial Bias in Facial Recognition Algorithms

Facial-recognition technology is already being used for applications ranging from unlocking phones to identifying potential criminals. Despite advances, it has still come under fire for racial bias: many algorithms that successfully identify white faces still fail to properly do so for people of color. Recently the National Institute of Standards and Technology (NIST) published a report showing how 189 face-recognition algorithms, submitted by 99 developers across the globe, fared at identifying people from different demographics.

Along with other findings, NIST’s tests revealed that many of these algorithms were 10 to 100 times more likely to inaccurately identify a photograph of a black or East Asian face, compared with a white one. In searching a database to find a given face, most of them picked incorrect images among black women at significantly higher rates than they did among other demographics. This report is the third part of the latest assessment to come out of a NIST program called the Face Recognition Vendor Test (FRVT), which assesses the capabilities of different face-recognition algorithms.

More information:

https://www.scientificamerican.com/article/how-nist-tested-facial-recognition-algorithms-for-racial-bias/