Deep Head Pose: Gaze-Direction Estimation in Multimodal Video

In this paper we present a convolutional neuralnetwork (CNN)-based model for human head pose estimation inlow-resolution multi-modal RGB-D data. We pose the problemas one of classification of human gazing direction. We furtherfine-tune a regressor based on the learned deep classifier. Next wecombine the two models (classification and regression) to estimateapproximate regression confidence. We present state-of-the-artresults in datasets that span the range of high-resolution humanrobot interaction (close up faces plus depth information) data tochallenging low resolution outdoor surveillance data. We buildupon our robust head-pose estimation and further introduce anew visual attention model to recover interaction with theenvironment. Using this probabilistic model, we show thatmany higher level scene understanding like human-human/sceneinteraction detection can be achieved. Our solution runs inreal-time on commercial hardware

Noise resilient discrete-time cross-relation sensor characterisation

In-situ characterisation of thermocouple sensors is a challenging problem. Recently the authors presented a blind characterisation technique based on the cross-relation method of blind identification. The method allows in-situ identification of two thermocouple probes, each with a different dynamic response, using only sampled sensor measurement data. While the technique offers certain advantages over alternative methods, including low estimation variance and the ability to compensate for noise induced bias, the robustness of the method is limited by the multimodal nature of the cost function. In this paper, a normalisation term is proposed which improves the convexity of
the cost function. Further, a normalisation and bias compensation hybrid approach is presented that exploits the advantages of both normalisation and bias compensation. It is found that the optimum of the hybrid cost function is less biased and more stable than when only normalisation is applied. All results were verified by simulation.

Novel temperature-activated, humidity-sensitive optical sensor

A novel, colorimetric, temperature-activated humidity indicator is presented, with a colour change based on the semi-reversible aggregation of thiazine dyes (esp. methylene blue, MB) encapsulated within the polymer, hydroxypropyl cellulose (HPC). The initially purple MB/HPC film is activated by heat treatment at 370 °C for 4 s, at which point the film (with a colour associated with a highly aggregated form of MB; λmax = 530 nm) becomes blue (indicating the presence of monomeric and dimeric MB; i.e. with λmax = 665; 605 nm respectively). The blue, heat-treated MB/HPC films respond to an ambient environment with a relative humidity (RH) exceeding 70% at 21 °C within seconds, returning to their initial purple colour. This colour change is irreversible until the film is heat-treated once more. When exposed to a lower RH of up to ca. 47%, the film is stable in its blue form. In contrast, a MB/HPC film treated only at 220 °C for 15 s also turns a blue colour and responds in the same way to a RH value of ca. 70%, but it is unstable at moderate RH 37-50% values, so that it gradually returns to its purple form over a period of approximately 6 hours. The possible use of the high heat-treated MB/HPC humidity indicator in the packaging of goods that cannot tolerate high RH, such as dry foods and electronics, is discussed.